Health Outcomes in Older Men with Localized Prostate Cancer: Results from the Prostate Cancer Outcomes Study

Article Outline

• Abstract
• Methods
  • Study Participants
  • Data Collection
  • Statistical Analysis
    • Propensity Score
    • Outcomes Models
• Results
• Discussion
• Acknowledgments
• References
• Copyright

Abstract 

Purpose

We compared health-related quality-of-life (HRQOL) outcomes and survival of men with localized prostate cancer who received aggressive treatment with those receiving conservative management.

Methods

We conducted a population-based cohort study of men aged 75 to 84 years when diagnosed with a clinically localized cancer in 1994 or 1995. We used medical record abstractions and patient surveys to obtain clinical and HRQOL data at diagnosis and 24-month follow-up. We used a propensity score method to adjust for baseline differences between men treated with radical prostatectomy or radiation therapy (n = 175) and men who received hormone therapy or no treatment (n = 290). Propensity scores were used in regression analyses comparing HRQOL outcomes between treatment groups. Overall and disease-specific survivals were estimated with multivariate proportional hazards models.

Results

At 24 months following diagnosis, aggressively treated men were more likely to report daily urinary leakage (odds ratio [OR] = 2.9, 95% confidence interval [CI] 1.2-7.0) and to be bothered by urinary problems (OR = 5.1, 95% CI, 1.3-9.1) and sexual problems (OR = 2.8, 95% CI, 1.2-6.3). The adjusted disease-specific mortality hazard ratio was 0.43 (95% CI, 0.15, 1.28), favoring aggressive treatment. However, the absolute 5-year disease-specific survival difference was only 6% (98% vs 92%). Over 80% of all deaths were from other causes.

Conclusions

Aggressive treatment was associated with significant decreases in disease-specific HRQOL. However, men who were aggressively treated for localized cancer had a minimally reduced absolute risk of dying from prostate cancer. Physicians and older patients should consider these outcomes in making decisions about screening and treatment.

Keywords:  Prostatic neoplasms , Prostatectomy , Radiotherapy , Outcome assessment , Aged

Among the many uncertainties surrounding prostate cancer is the appropriate age to stop screening. Guidelines suggest that screening is unlikely to benefit men with a life expectancy of fewer than 10 years.¹, ², ³ This observation reflects the often-indolent course of prostate cancer,⁴ the potential harms from aggressive treatments such as radical prostatectomy and radiation therapy,⁵, ⁶ and the high likelihood that older men will die from competing morbidities.⁷, ⁸ Based on actuarial tables, the upper age for screening a man in average health would be 75 years, although older men in excellent health may have a similar life expectancy.⁹ A recent decision analysis concluded that aggressively treating a healthy 75-year-old man with a poorly differentiated cancer could increase quality-adjusted life expectancy by about 1 year.¹⁰

Population- and practice-based surveys show high rates of prostate cancer screening in men aged 75 and older,¹¹, ¹², ¹³ and the incidence of early-stage cancers in older men increased markedly following the advent of prostate-specific antigen (PSA) testing.¹⁴ Concomitantly, their rates of aggressive treatment increased as well, even though treatment morbidity and mortality are substantially higher in older patients.¹⁵, ¹⁶, ¹⁷ These data raise questions about the appropriateness of screening and treating older patients, particularly because the effects on survival and health-related quality of life (HRQOL) are not well characterized. The major randomized screening¹⁸, ¹⁹ and treatment²⁰, ²¹ trials have consistently excluded older men.

We used data from the population-based, observational Prostate Cancer Outcomes Study to describe the baseline demographic, socioeconomic, and clinical characteristics of men who were aged 75 to 84 years when diagnosed with a clinically localized prostate cancer. Using a propensity score method to control for factors related to treatment selection, we compared disease-specific and general HRQOL outcomes and survival between the men who received aggressive treatment (radical prostatectomy or radiation therapy) and those who were conservatively managed (hormone therapy or no treatment).

Back to Article Outline

Methods 

Study Participants 

The Prostate Cancer Outcomes Study (PCOS) is based on an existing tumor registry system, the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) program. Details of the PCOS have been published elsewhere.²² Briefly, the study identified men diagnosed with prostate cancer between October 1, 1994 and October 31, 1995. Patients were diagnosed in six SEER tumor registries covering the states of Connecticut, Utah, and New Mexico, and the metropolitan areas of Atlanta, Georgia; Los Angeles County, California; and King County, Washington. Participating subjects provided informed consent, and the Institutional Review Board of each PCOS site approved the study.

The PCOS randomly sampled, within strata of age, race/ethnicity, and tumor registry, a total of 5672 subjects from the 11137 eligible prostate cancer cases. Among the sampled subjects, 3533 (62%) completed HRQOL survey questionnaires 6 or 12 months after initial diagnosis, including 3073 men with clinically localized prostate cancer. The current analyses used data collected from the 465 men aged 75 to 84 years at diagnosis.

Data Collection 

All subjects consented to a review of their medical records relating to care for prostate cancer. Abstractors collected information on demographic factors, clinical symptoms, diagnostic examinations, biopsy results, tumor characteristics (PSA level and Gleason score), clinical staging, and treatment.

We surveyed subjects approximately 6 months after diagnosis to collect data on disease-specific HRQOL, using a prostate-cancer specific instrument based on items from 3 existing instruments.²³, ²⁴, ²⁵ Subjects reported on their current and baseline (just before diagnosis) urinary, bowel, and sexual functions and perceived bother with these functions. The functional (as opposed to bother) items in each category were combined to create a composite score ranging from 0 to 100 points. General HRQOL for the previous 4 weeks was measured using six domains of the Medical Outcomes Study SF-36: general health, physical problems, emotional problems, bodily pain, vitality, and mental health.²⁶ Domain scores were transformed to a 0-100 point scale. Higher scores represented better function.

Survey items included race/ethnicity, employment status, educational level, household income, insurance coverage, and marital status. We asked about 12 medical conditions that could affect prostate cancer treatment decisions and long-term quality of life. The conditions were derived from the Charlson index²⁷ and modified by the PCOS investigators. Subjects reporting cerebrovascular disease, inflammatory bowel disease, liver disease, or gastric ulcers received one point on his comorbidity score for each condition. Subjects reporting arthritis, diabetes, depression, hypertension, chest pain, heart attack, heart failure, or chronic lung disease that limited his activity or required prescription medications received an additional point.

We identified subjects with clinically localized prostate cancer based on information abstracted from medical records. We defined localized tumors as confined to the prostate by digital rectal examination with no positive scans (magnetic resonance imaging, computed tomography, bone scan) or evidence of metastases. We defined aggressive treatment as receiving radical prostatectomy or radiation therapy within 12 months after diagnosis.

Subjects were surveyed 24 months after diagnosis regarding general and disease-specific HRQOL. Subjects also were asked about the overall effects of prostate cancer or treatment on physical discomfort, health worries, activity limitations, overall bother, treatment satisfaction, and willingness to make the same treatment decision again. We ascertained vital status through 84 months following diagnosis using SEER data.

Statistical Analysis 

Older men receiving aggressive treatment for clinically localized prostate cancer have different characteristics than men receiving conservative management, and these characteristics also may be associated with disease outcomes. Therefore, we used a propensity score method to adjust treatment outcomes for this potential selection bias.²⁸ We first evaluated the baseline (at time of diagnosis) patient characteristics thought to predict receiving aggressive therapy, based on expert opinions and literature review. We then estimated the propensity score for receiving aggressive treatment by including the relevant baseline characteristics in a single multivariable logistic regression analysis with treatment intensity (aggressive vs. conservative) as the dependent variable. The predicted values of the regression model estimated the probability (propensity) of receiving aggressive treatment for each patient conditional on the values of his baseline characteristics.

We evaluated whether the propensity score could be used as a covariate in regression analyses. We first estimated the Wald F-statistic and P value for the patient characteristics associated with treatment before and after adjustment for the propensity score. We then evaluated associations between treatment and patient characteristics within one or more quintiles of the propensity score by adding interaction terms to the model. Finally, we inspected the distribution of the patient characteristics within treatment categories stratified by propensity score quintile.

We used multivariable logistic regression analyses to compare aggressively treated and conservatively managed patients with respect to urinary, bowel, and sexual domain items ascertained on the 24-month survey. We included the following covariates in all multivariate models: treatment propensity score, age at diagnosis, race/ethnicity, educational attainment, comorbidity score, and the baseline response for the function or bother item. We estimated adjusted odds ratios as the measure of treatment effect, with conservative management as the reference group. We estimated predicted marginal proportions of the outcome for each treatment group that were directly standardized to the distribution of the covariates among the entire weighted sample.²⁹ Similar logistic models were used to evaluate items assessing the overall effects of prostate cancer and treatment, treatment satisfaction, and willingness to undergo the same treatment again. We compared general HRQOL outcomes, as measured by the SF-36 on the 24-month survey, and urinary, bowel, and sexual functional scores between the two treatment groups using linear regression models.

We used the log-rank test to determine statistical significance of estimated overall and disease-specific survival differences for the 2 treatment groups. We used a Cox proportional hazards model to estimate the mortality hazard ratio.

The study used a stratified sampling strategy defined by age, race/ethnicity, and region. With the exception of some descriptive statistics, all analyses have taken this sampling into account using the statistical package SUDAAN.³⁰ We used the inverse of the sample proportion for weights (the Horvitz-Thompson weight) and specified a stratified with-replacement study design. Variances were estimated using Taylor linearization methods. All significance tests were 2-sided, and a P value <.05 was considered statistically significant. SAS Version 8.02 was used to prepare the data for input into the SUDAAN procedures and for some descriptive statistics.³¹

Back to Article Outline

Results 

A total of 465 men, aged 75 to 84 years, were diagnosed with a clinically localized prostate cancer. Baseline demographic, socioeconomic, and clinical characteristics of the cohort are shown in Table 1. The majority of men were white, married, and had at least a high school education. Nearly 80% of subjects had a moderately or poorly differentiated cancer, and 81% had PSA levels greater than 4.0 ng/mL. Few men had multiple comorbidities, and most reported no baseline problems with urinary or bowel function. However, 29% reported having moderate to big problems with sexual function. Over 79% of subjects reported that they discussed aggressive treatment, either radical prostatectomy or radiation therapy, with a physician. One hundred seventy-five men (38%) received aggressive treatment (38 radical prostatectomy, 137 radiation therapy), whereas 290 (62%) were conservatively managed (116 hormone therapy, 174 no active treatment).

Table 1. Baseline Characteristics of Subjects Aged 75 to 84 Years with Clinically Localized Prostate Cancer^⁎

Characteristic	Number (%)†
Age at diagnosis
75-79	327(70.2)
80-84	138(29.8)
Race
Non-Hispanic white	364(84.2)
Non-Hispanic black	47(7.7)
Hispanic	54(8.0)
Tumor registry
Connecticut	97(22.9)
New Mexico	53(13.6)
Seattle	72(7.9)
Utah	95(10.5)
Atlanta	28(11.3)
Los Angeles	120(33.8)
Current marital status
Married	343(72.6)
Not married	116(26.1)
Sexual partner
Yes	308(64.4)
No	130(29.5)
Educational attainment
Less than high school graduate	144(29.9)
High school graduate/some college	205(43.3)
College graduate	109(25.3)
Income
≤$20,000	148(30.8)
$20,000 - $40,000	158(31.0)
≥$40,000	76(18.3)
Insurance
Private	364(79.6)
Medicare/Public	58(12.5)
Gleason score
2 to 4	68(12.6)
5	74(14.7)
6	115(25.9)
7	109(25.7)
8 to 10	55(10.8)
PSA (ng/mL)
<4	34(5.9)
4 to 10	147(32.1)
>10 to 20	116(26.2)
≥20	110(23.4)
Digital rectal examination
Abnormal	246(53.8)
Normal	143(30.6)
Comorbidity score
0	135(30.4)
1	164(32.7)
2	83(20.0)
3+	83(16.8)
History of other cancers
Yes	31(7.5)
No	434(92.5)
Baseline incontinence problem
No problem	304(67.2)
Small or very small problem	111(22.0)
Moderate or big problem	38(8.3)
Mean urinary function score	86.1
Baseline bowel function problem
No problem	328(71.8)
Small or very small problem	107(21.8)
Moderate or big problem	22(5.1)
Mean bowel function score	92.7
Baseline sexual function problem
No problem	181(39.0)
Small or very small problem	113(24.5)
Moderate or big problem	136(28.6)
Mean sexual function score	39.7

We found considerable baseline differences between the treatment groups (Table 2). Aggressively treated men were characterized by younger age, white race, being married, higher income, a lower PSA level, a lower prevalence of chronic lung disease or heart attack, and being less bothered by urinary or sexual problems. However, we found no statistically significant associations between patient characteristics and treatment after adjustment for the propensity score.

Table 2. Baseline Characteristics Significantly Associated with Aggressive Treatment for Clinically Localized Prostate Cancer in Men Aged 75 to 84 Years, Before and After Adjustment for Propensity Scores

Variable	n	Received aggressive treatment (%)	Unadjusted Wald F (P value)	Adjusted for propensity score Wald F (P value)
All subjects	465	37.7	–	–
Age at diagnosis (years)			19.48(.0000)	0.34(.56)
75-79	327	45.6
80-84	138	19.0
Race/ethnicity			5.30(.005)	0.12(.88)
Non-Hispanic White	364	40.5
Non-Hispanic Black	47	19.9
Hispanic	54	24.2
Tumor registry			2.34(.04)	0.04(.99)
Connecticut	97	41.2
New Mexico	53	30.0
Seattle	72	38.9
Utah	95	50.5
Atlanta	28	52.0
Los Angeles	120	29.2
Current marital status			3.54(.03)	0.02(.98)
Married	343	42.0
Not married	116	26.3
Income			5.85(.0006)	0.05(.99)
<$20000	148	27.9
$20000-40000	158	42.1
>$40000	76	58.1
PSA (ng/mL)			3.76(.005)	0.04(.99)
<4	34	29.8
4 to 10	147	45.9
>10 to 20	116	46.6
>20	110	24.9
Comorbidities
Chronic lung disease	27	17.3	4.10(.04)	0.01(.92)
Myocardial infarction	57	22.5	5.15(.02)	0.03(.87)
Incontinence problem			3.22(.02)	0.16(.93)
No problem	304	39.9
Small/very small	111	36.2
Moderate/big	38	16.1
Urinary frequency			2.93(.02)	0.04(.99)
Rarely/not at all	234	39.4
< Half the time	86	36.4
Half the time	76	50.6
> Half the time	50	19.1
Sexual function problem			5.10(.0018)	0.15(.93)
None	181	36.7
Small/very small	113	46.7
Moderate/big	136	39.9

Three hundred forty-four subjects (74.0%) completed the 24-month follow-up survey, including 143 (81.7%) of those who received aggressive treatment and 201 (69.3%) of those who were conservatively managed. Table 3 compares the 24-month outcomes for the disease-specific functional and bother items between treatment groups. Men undergoing aggressive treatment were significantly more likely to report daily urinary leakage and to be bothered by urinary incontinence. They were also significantly more likely to report frequent bowel movements and were twice as likely to be bothered by bowel problems, although this was not statistically significant. Sexual function was poor for both treatment groups, but men who received aggressive treatment were significantly more likely be bothered by sexual dysfunction.

Table 3. Comparison of 24-Month Survey Responders on Urinary, Bowel, and Sexual Domain Items^⁎

Domain and survey items	Conservative %	Aggressive %	Odds ratio (95% CI)†	P value
Urinary
No control or frequent leakage	6.7	12.4	2.0(0.7-5.7)	.17
Leaks ≥ once daily	10.1	22.2	2.9(1.2-7.0)	.015
Frequent urination > half the time	10.3	7.9	0.7(0.3-1.7)	.46
Bothered by dripping or leaking urine	4.2	14.4	5.1(1.3-19.1)	.02
Bowel
Frequency some/almost all days	17.7	29.8	2.3(1.0-5.0)	.04
Bowel urgency some/almost all days	20.3	27.6	1.7(0.8-3.5)	.19
Painful bowel movements some/almost all days	11.7	16.8	1.6(0.7-3.7)	.24
Bothered by bowel function problems	4.4	9.4	2.4(0.8-7.5)	.12
Sexual
No/little interest in sexual activity	66.3	64.4	0.9(0.4-1.8)	.75
No sexual activity	73.3	68.1	0.7(0.3-2.3)	.63
Erections not firm enough for intercourse	88.0	80.1	0.4(0.1-1.4)	.16
No erections/a lot of difficulty keeping erections	86.0	83.2	0.4(0.3-2.3)	.63
Bothered by sexual function problems	23.5	43.3	2.8(1.2-6.3)	.01

Table 4 shows the changes in mean composite scores for urinary, bowel, and sexual function from diagnosis to 24-month follow-up. Overall, urinary function decreased significantly more for men receiving aggressive treatment than with conservative management. Although bowel function remained relatively high for both groups, the decrease for men receiving aggressive treatment was significantly greater. Sexual function decreased substantially for both groups. For all three domains, the functional decrease over time was greater for those men with higher baseline function (P <.01 for each domain). The pattern of changes suggests that the treatment effect varied by the baseline function level, although the interaction was significant only for bowel function, P <.01.

Table 4. Mean Baseline Function Score and Change from Baseline at 24-Months Follow-up, Stratified by Treatment Group and Baseline Function Level^⁎

Men who received aggressive treatment were more likely to report being delighted or very pleased with their treatment decision, although both groups were equally likely to report willingness to make the same treatment decision again (Table 5). We found no treatment differences for other outcomes related to prostate cancer or treatment. However, two domains of general health-related quality of life, as measured by the SF-36, differed by treatment. Men who received aggressive treatment had significantly higher scores in the general health (66.8 vs 59.2, P <.04) and physical problems (60.8 vs 47.7, P <.04) scales than those managed conservatively.

Table 5. Overall Effects of Prostate Cancer and Treatment^⁎

Outcome	Conservative (%)	Aggressive (%)	Odds ratio (95% CI)†	Wald F (P value)
Treatment satisfaction	52.8	68.1	2.1(1.0-4.4)	3.74(.05)
Delighted/very pleased vs mostly
satisfied/mixed/dissatisfied
Make same treatment decision again if given chance	56.5	53.3	0.9(0.4-1.7)	0.15(.70)
Definitely yes vs probably yes or not
Physical discomfort related to cancer/treatment	13.9	20.6	1.8(0.8-4.4)	1.78(.18)
A lot/some vs only a little/none at all
Health worries related to cancer/treatment	19.7	18.0	0.9(0.4-1.9)	0.10(.75)
A lot/some vs only a little/none at all
Limited daily activities related to cancer/treatment	12.1	13.8	1.2(0.4-3.8)	0.13(.72)
A lot/some vs only a little/none at all
Overall bother related to cancer/treatment	19.3	16.5	0.8(0.3-1.9)	0.24(.62)
A lot/some vs only a little/none at all

By the end of follow-up, 18 deaths (5 from prostate cancer) occurred in the aggressive treatment group, compared with 137 deaths (23 from prostate cancer) in the conservatively managed group. The estimated 5-year overall survival was 82% (95% confidence interval [CI] 0.77-0.88) in men who received aggressive treatment, versus 63% (95% CI, 0.57-0.68) for conservative management, log-rank P <.0001. The adjusted hazard ratio for overall mortality was 0.67 (95% CI, 0.44-1.03), favoring aggressive treatment. Men who received aggressive treatment had an estimated 5-year prostate cancer-specific survival of 98% (95% CI, 0.95-1.00) versus 92% (95% CI, 0.88-0.96) for conservative management, log rank P = .01. The adjusted hazard ratio for prostate-cancer specific mortality was 0.43 (95% CI, 0.15-1.28), favoring aggressive treatment.

We found significant differences between men who completed the 24-month survey and the non-responders. Responders were more likely to be white, married, college educated, more affluent, and covered by private insurance (data not shown). Completion rates also were higher for men with PSA levels below 10 ng/mL. Men who received aggressive treatment were more likely to complete the follow-up survey, but the difference was not significant, P = .18.

Back to Article Outline

Discussion 

Screening and treatment decisions for older men are challenging. A recent decision analysis suggested that healthy older men with high-grade tumors might gain quality adjusted life years with aggressive treatment.¹⁰ Prostate cancer decision models, however, have important limitations because there are no outcomes data from randomized screening or treatment trials enrolling older men. Furthermore, outcomes data from observational studies are often difficult to interpret due to selection bias. Decisions about aggressive treatment must balance the possible increases in life expectancy—which is already diminished in older men with comorbidities—with the more certain, immediate, and persistent treatment complications that can adversely affect HRQOL. Guidelines from professional organizations have suggested an upper age of 75 years for offering screening.¹, ², ³, ³²

We used longitudinal data from a population-based study of older men with localized prostate cancer to compare HRQOL outcomes and survival following aggressive treatment versus conservative management. Over one third of subjects received either radical prostatectomy or radiation therapy. Aggressively treated men were more likely to be bothered by urinary problems 24 months after diagnosis. Urinary function also decreased more following aggressive treatment, especially for men with the highest baseline function. Men receiving aggressive treatment had more problems with frequent bowel movements and a greater decrease in bowel function. Baseline sexual function was decreased for both groups; however, among aggressively treated men with the highest baseline sexual function, function decreased by 39% over 24 months. Although these men actually experienced a lesser decrease than those managed conservatively (who were older and often receiving hormone therapy), more reported being bothered by sexual dysfunction, 43% versus 24%. Their high baseline function might unrealistically have raised expectations for retaining erectile function.

The two treatment groups had similar levels of physical discomfort, health worries, activity limitations, and overall bother from prostate cancer or treatment. However, men receiving aggressive treatment had higher scores for the SF-36 general health and role-physical domains at 24 months follow-up. We would not expect aggressive treatment to improve these domain scores. Of note, we first measured these domains 6 months after diagnosis; men receiving aggressive treatment probably had higher baseline scores.

Although disease-specific survival was somewhat higher following aggressive treatment, few deaths among the entire study cohort were attributed to prostate cancer (18%), and the absolute 5-year survival difference was only 6%. Other investigators also have shown that older men with localized cancers were much more likely to die from competing morbidities, even in the absence of treatment.⁷, ³³ The major threat to validity in an observational outcomes study is selection bias. The difference in overall survival suggested that the propensity score analysis did not completely control for this bias, implying residual confounding. Nonetheless, men who received aggressive treatment still had poorer disease-specific HRQOL outcomes at follow-up.

Attrition bias was another potential limitation. Although 24% of the cohort did not complete the 24-month survey, mortality was 10%. Nonwhites and those of lower socioeconomic status were less likely to complete the follow-up survey. Our results may be less generalizable to these populations. We did not see a differential loss to follow-up in regard to treatment group, although nonrespondents had a significantly lower mean propensity score (data not shown). Given that a substantial proportion of nonrespondents had died, we attributed the lower propensity score to their poor health.

In summary, we found that aggressively treated older men with clinically localized prostate cancer were healthier at baseline and had more favorable tumor characteristics than those who were managed conservatively. Aggressive treatment minimally reduced the risk of dying from prostate cancer; disease-specific survival, however, was relatively high in both groups because most deaths were from other causes. However, by 24 months following diagnosis, men who received aggressive treatment had suffered more urinary and bowel dysfunction and were more bothered by sexual dysfunction. General health and physical function were higher for aggressively treated men, but this was likely due to residual selection bias. In the absence of any randomized trial data, our results support guidelines suggesting that men 75 years and older may not benefit from prostate cancer screening.¹, ², ³, ³³

Back to Article Outline

Acknowledgments 

This work was supported by Public Health Services contracts N01-PC-67007, N01-PCN-67009, N01-PC-67010, N01-PC-67006, N01-PC-67005, N01-PC-67000 from the National Cancer Institute, National Institutes of Health, Department of Health and Human Services, and the New Mexico VA Health Care System. We thank the men who participated in the Prostate Cancer Outcomes Study and their physicians. We thank the study teams at each of the research centers for their contributions. We also thank Arnold L. Potosky, PhD, for supporting this analysis.

Back to Article Outline

References 

1. Harris R , Lohr KN . Screening for prostate cancer (an update of the evidence for the U.S. Preventive Services Task Force) . Ann Intern Med . 2002;137:917–929
   • View In Article
2. American Urological Association . Prostate-specific antigen (PSA) best practice policy . Oncology . 2000;14:267–272 277-280
   • View In Article
3. Smith RA , Mettlin CJ , Davis KJ , Eyre H . American Cancer Society guidelines for the early detection of cancer . CA Cancer J Clin . 2000;50:34–49
   • View In Article
   • MEDLINE
   • CrossRef
4. Coley CM , Barry MJ , Fleming C , Mulley AG . Early detection of prostate cancer. Part I: Prior probability and effectiveness of tests. The American College of Physicians . Ann Intern Med . 1997;126:394–406
   • View In Article
   • MEDLINE
5. Stanford JL , Feng Z , Hamilton AS , et al.   Urinary and sexual function after radical prostatectomy for clinically localized prostate cancer (the Prostate Cancer Outcomes Study) . JAMA . 2000;283:354–360
   • View In Article
   • MEDLINE
   • CrossRef
6. Hamilton AS , Stanford JL , Gilliland FD , et al.   Health outcomes after external-beam radiation therapy for clinically localized prostate cancer (results from the Prostate Cancer Outcomes Study) . J Clin Oncol . 2001;19:2517–2526
   • View In Article
   • MEDLINE
7. Johansson JE , Andren O , Andersson SO , et al.   Natural history of early, localized prostate cancer . JAMA . 2004;291:2713–2719
   • View In Article
   • CrossRef
8. Albertsen PC , Fryback DG , Storer BE , Kolon TF , Fine J . The impact of co-morbidity on life expectancy among men with localized prostate cancer . J Urol . 1996;156:127–132
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (738 KB)
   • CrossRef
9. Welch HG , Albertsen PC , Nease RF , Bubolz TA , Wasson JH . Estimating treatment benefits for the elderly (the effect of competing risks) . Ann Intern Med . 1996;124:577–584
   • View In Article
   • MEDLINE
10. Alibhai SM , Naglie G , Nam R , Trachtenberg J , Krahn MD . Do older men benefit from curative therapy of localized prostate cancer? . J Clin Oncol . 2003;21:3318–3327
    • View In Article
    • MEDLINE
    • CrossRef
11. Lu-Yao G , Stukel TA , Yao SL . Prostate-specific antigen screening in elderly men . J Natl Cancer Inst . 2003;95:1792–1797
    • View In Article
12. McNaughton Collins M , Stafford RS , Barry MJ . Age-specific patterns of prostate-specific antigen testing among primary care physician visits . J Fam Pract . 2000;49:169–172
    • View In Article
    • MEDLINE
13. Ross LE , Coates RJ , Breen N , Uhler RJ , Potosky AL , Blackman D . Prostate-specific antigen test use reported in the 2000 National Health Interview Survey . Prev Med . 2004;38:732–744
    • View In Article
    • MEDLINE
    • CrossRef
14. Ries LAG , Eisner MP , Kosary CL , et al.   SEER Cancer Statistics Review, 1973-1999 . Bethesda, MD: National Cancer Institute; 2002;
    • View In Article
15. Lu-Yao GL , McLerran D , Wasson J , Wennberg JE  The Prostate Patient Outcomes Research Team . An assessment of radical prostatectomy. Time trends, geographic variation, and outcomes . JAMA . 1993;269:2633–2636
    • View In Article
    • MEDLINE
16. Bubolz T , Wasson JH , Lu-Yao G , Barry MJ . Treatments for prostate cancer in older men (1984-1997) . Urology . 2001;58:977–982
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (363 KB)
    • CrossRef
17. Wilt TJ , Cowper DC , Gammack JK , Going DR , Nugent S , Borowsky SJ . An evaluation of radical prostatectomy at Veterans Affairs Medical Centers (time trends and geographic variation in utilization and outcomes) . Med Care . 1999;37:1046–1056
    • View In Article
    • MEDLINE
    • CrossRef
18. Gohagan JK , Prorok PC , Hayes RB , Kramer BS . The Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial of the National Cancer Institute (history, organization, and status) . Control Clin Trials . 2000;21(6 Suppl):251S–272S
    • View In Article
    • Abstract
    • Full-Text PDF (908 KB)
    • CrossRef
19. Schroder FH . The European Screening Study for Prostate Cancer . Can J Oncol . 1994;4(Suppl 1):102–105 discussion 106-109
    • View In Article
20. Holmberg L , Bill-Axelson A , Helgesen F , et al.   A randomized trial comparing radical prostatectomy with watchful waiting in early prostate cancer . N Engl J Med . 2002;347:781–789
    • View In Article
    • CrossRef
21. Wilt TJ , Brawer MK . The Prostate Cancer Intervention Versus Observation Trial (a randomized trial comparing radical prostatectomy versus expectant management for the treatment of clinically localized prostate cancer) . J Urol . 1994;152:1910–1914
    • View In Article
    • MEDLINE
22. Potosky AL , Knopf K , Clegg LX , et al.   Quality-of-life outcomes after primary androgen deprivation therapy (results from the Prostate Cancer Outcomes Study) . J Clin Oncol . 2001;19:3750–3757
    • View In Article
    • MEDLINE
23. Fowler FJJ , Barry MJ , Lu-Yao G , Roman A , Wasson J , Wennberg JE . Patient-reported complications and follow-up treatment after radical prostatectomy. The National Medicare Experience: 1988-1990 (updated June 1993) . Urology . 1993;42:622–629
    • View In Article
    • Abstract
    • Full-Text PDF (881 KB)
    • CrossRef
24. Litwin MS , Hays RD , Fink A , Ganz PA , Leake B , Brook RH . The UCLA Prostate Cancer Index (development, reliability, and validity of a health-related quality of life measure) . Med Care . 1998;36:1002–1012
    • View In Article
    • MEDLINE
    • CrossRef
25. Talcott JA , Rieker P , Clark JA , et al.   Patient-reported symptoms after primary therapy for early prostate cancer (results of a prospective cohort study) . J Clin Oncol . 1998;16:275–283
    • View In Article
    • MEDLINE
26. Ware JE , Sherbourne CD . The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection . Med Care . 1992;30:473–483
    • View In Article
    • MEDLINE
    • CrossRef
27. Charlson ME , Pompei P , Ales KL , MacKenzie CR . A new method of classifying prognostic comorbidity in longitudinal studies (development and validation) . J Chronic Dis . 1987;40:373–383
    • View In Article
    • MEDLINE
    • CrossRef
28. Rubin DB . Estimating causal effects from large data sets using propensity scores . Ann Intern Med . 1997;127:757–763
    • View In Article
    • MEDLINE
29. Graubard BI , Korn EL . Predictive margins with survey data . Biometrics . 1999;55:652–659
    • View In Article
    • MEDLINE
    • CrossRef
30. Shah BV , Barnwell BG , Bieler GS . SUDAAN User’s Manual, Release 7.5 ed . Research Triangle Park, NC: Research Triangle Institute; 1997;
    • View In Article
31. SAS/STAT User’s Guide, Version 8 . Cary, North Carolina: SAS Institute, Inc; 2001; 1999-
    • View In Article
32. American College of Physicians . Screening for prostate cancer . Ann Intern Med . 1997;126:480–484
    • View In Article
    • MEDLINE
33. Albertsen PC , Hanley JA , Gleason DF , Barry MJ . Competing risk analysis of men aged 55 to 74 years at diagnosis managed conservatively for clinically localized prostate cancer . JAMA . 1998;280:975–980
    • View In Article
    • MEDLINE
    • CrossRef