This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Litwin, M. S. Articles by McGuigan, K. A. Search for Related Content PubMed PubMed Citation Articles by Litwin, M. S. Articles by McGuigan, K. A.

Accuracy of Recall in Health-Related Quality-of-Life Assessment Among Men Treated for Prostate Cancer

From the Departments of Urology and Health Services, University of California Los Angeles, Los Angeles, and RAND Health Program, Santa Monica, CA.

Address reprint requests to Mark S. Litwin, MD, MPH, University of California Los Angeles Department of Urology, Box 951738, Los Angeles, CA 90095-1738; email mlitwin{at}ucla.edu

	ABSTRACT

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION APPENDIX: UCLA Prostate Cancer... REFERENCES

 
PURPOSE: To determine the accuracy of patient recall of health-related quality of life (HRQOL) in men who have undergone radical prostatectomy for early-stage prostate cancer.

PATIENTS AND METHODS: Patients enrolled onto a longitudinal, observational cohort study of HRQOL after radical prostatectomy for early-stage prostate cancer were asked to assess their baseline HRQOL before surgery. They were later asked to recall their baseline HRQOL at intervals of 7 to 37 months after surgery. The two views of baseline HRQOL (actual and recall) were compared. HRQOL was measured with established instruments (the RAND 12-Item Short-Form Health Survey and a validated short form of the University of California Los Angeles Prostate Cancer Index) that addressed impairment in the physical, mental, urinary, bowel, and sexual domains.

RESULTS: Overall, recall was poor. Patients tended to remember their baseline HRQOL as being better than it actually was. This effect was particularly striking for urinary and sexual function. Greater education and younger age diminished this effect in some domains. The effect did not vary with time since surgery.

CONCLUSION: Men undergoing radical prostatectomy for early-stage prostate cancer do not accurately recall their pretreatment HRQOL when asked several months or years later. This recall bias is constant throughout a period of 6 months to 3 years after surgery. By collecting data before treatment and observing subjects longitudinally, investigators can ensure that HRQOL changes are analyzed in the context of any impairment that may have been present at baseline. If a longitudinal study is not feasible, then great caution must be used if patients are asked to recall their pretreatment HRQOL.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION APPENDIX: UCLA Prostate Cancer... REFERENCES

 
ADENOCARCINOMA OF the prostate, the most common noncutaneous malignancy in American men,¹ presents patients and physicians with many challenges when making treatment decisions. Although the traditional goal in cancer care has been to maximize the length of survival, the tumor's relatively indolent course in some patients may obviate the need for curative therapy. In recent years, patients and physicians have grown increasingly interested in also optimizing the quality of life. Treatment options for localized disease include surgery (radical prostatectomy), radiotherapy (external-beam or brachytherapy), expectant management, and a variety of newer approaches such as cryotherapy. Each of these approaches is associated with a distinct set of potential risks and benefits. Each therapy can significantly impact both quantity and quality of life.

The most common complications of interventions for prostate cancer include impotence, incontinence, and proctitis. Although alterations in the sexual, urinary, and bowel domains may be mild or severe, they continue to affect men's lives long after treatment is complete. However, older men with or without prostate cancer typically do not have perfect sexual, urinary, and bowel function at baseline.^2-7 For example, in a large sample of patients who were treated expectantly for early-stage prostate cancer, sexual dysfunction rates approximated those in patients who had undergone radiotherapy, and both groups functioned at levels that were significantly lower than those of age-matched controls.⁸ General quality of life may also decline as patients grow older. Hence, patients may already be experiencing age- or tumor-related quality-of-life impairment in the general or pelvic domains even before any therapy is undertaken.

When studying the effects of surgery or radiation on the quality of life of men with early-stage prostate cancer, it is critical to obtain baseline measurements before treatment. Then, longitudinal assessments can be used by clinicians to track changes in patients' quality of life after treatment and to compare these outcomes with pretreatment status. It is important to have a measure of baseline functioning because patients with localized prostate cancer may experience impairments even before treatment. Ideally, baseline assessment should be measured before treatment; however, prospective assessment is not always feasible. For example, when registries are used to accrue patients for quality-of-life studies, a time lag of up to 6 months (for hospital registries) or 1 to 2 years (for central cancer registries) between diagnosis and appearance in the registry is common. Because most patients with early-stage prostate cancer begin treatment within several weeks of diagnosis, cases drawn from these registries would become available for assessment well after their initial treatment.

Without prospective, pretreatment measurements, the remaining options are to use an age-matched healthy control group or to assess baseline quality of life with recall reporting. However, retrospective reports of quality of life and health status may not provide valid measures of actual functioning because their accuracy depends on human factors, such as memory and perception, that are inherently subject to distortion.⁹

The accuracy of recall reporting to assess pretreatment quality of life in men with early-stage prostate cancer is presently unknown. We undertook this study to ascertain how well men could recall their pretreatment quality of life. We used an established, longitudinal cohort of men managed with radical prostatectomy to compare recall reporting of baseline quality of life with actual baseline measurements of the same variables. We sought to assess overall accuracy and to determine whether agreement between retrospective recall and actual baseline reports varies as a function of characteristics such as age, education, or time since surgery.

	PATIENTS AND METHODS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION APPENDIX: UCLA Prostate Cancer... REFERENCES

 
Subjects
Patients were drawn from an ongoing longitudinal, observational database of men who underwent radical prostatectomy as primary therapy for clinically localized prostate cancer. The database captured health-related quality of life (HRQOL) at baseline before treatment, at 3-month intervals during the year after surgery, then at 6-month intervals until 5 years after surgery.¹⁰ General and disease-specific HRQOL were measured with established, validated instruments confidentially self-administered with the approval of the University of California Los Angeles (UCLA) Human Subjects Protection Committee. All database patients who had been observed for at least 6 months after radical prostatectomy were eligible for the present study. Of these, a 50% random probability sample (123 patients) was selected to receive the recall survey.

Recall Survey
We mailed the recall survey to each patient in the sample along with a cover letter and a prestamped return envelope and then used telephone reminders for subjects who did not respond within 2 weeks. The recall survey asked patients to think back to the 4 weeks before their prostate surgery and recall the various aspects of their quality of life in physical, emotional, social, and functional terms. To minimize response burden, the recall survey contained only 26 items (see below). We then compared patients' responses to the recall survey with their responses to the baseline survey, which had been administered just before radical prostatectomy, regarding the 4 weeks before surgery.

Outcome Measures
Actual baseline general HRQOL was assessed with the RAND 36-Item Health Survey 1.0 (short-form [SF]-36) (RAND, Santa Monica, CA).¹¹ The survey is a self-administered, 36-item questionnaire that quantifies general HRQOL with eight multi-item scales and two summary scales, the physical component summary scale and the mental component summary scale. The scales are scored from 0 to 100, with higher scores representing better health states. The RAND 36-Item Health Survey has been thoroughly tested and has been shown to be highly reliable and valid in a variety of populations, including men with prostate cancer.^8,12,13 The survey has been widely used to measure general HRQOL in many types of patients.

Actual baseline disease-specific HRQOL was assessed with the UCLA Prostate Cancer Index (PCI).¹⁴ It is a self-administered 20-item questionnaire that quantifies prostate-specific HRQOL in six domains, urinary function and bother, bowel function and bother, and sexual function and bother. The urinary, bowel, and sexual function scales focus on incontinence, proctitis, and sexual difficulties, respectively, whereas the bother scales focus on how much the patient is troubled by the dysfunction. The six scales are scored from 0 to 100, with higher scores representing better health states. The UCLA PCI has been shown to be reliable and valid in populations of older men with early- and late-stage prostate cancer and in older men without prostate cancer.^8,10,15

Recalled HRQOL was measured with the short forms of each instrument. Recalled general HRQOL was assessed with the SF-12, a 12-item version of the SF-36, which has been shown to be reliable and valid and highly correlates with the SF-36.^16,17 The SF-12 quantifies HRQOL with the physical component summary scale and mental component summary scale rather than with the eight domain scores.

Recalled prostate-targeted HRQOL was assessed with a 14-item short form of the UCLA PCI (PCI-SF) validated specifically for this study. To create the PCI-SF, we used data from the original PCI validation sample¹⁴ to eliminate duplicative items while maintaining excellent reliability and validity. This was accomplished by removing two of five items from the urinary function scale, one of four items from the bowel function scale, and three of eight items from the sexual function scale. The three bother scales are single-item scales and were left intact. The PCI-SF is presented in the Appendix.

Statistical Analysis
We measured the proportion of variation in the original PCI subscales accounted for by the short-form versions to assess construct validity. We used Cronbach's alpha¹⁸ to calculate internal consistency reliability for the general and prostate-targeted HRQOL scales. Student's t test and Fisher's exact test were used to assess differences between respondents and nonrespondents on sociodemographic characteristics and actual baseline HRQOL measures. Differences between baseline and recall assessments were evaluated using paired t tests. Analysis of covariance and difference score regressions were used to examine variations in recall accuracy using patient characteristics (age, time since baseline, and education). Significant mean differences in recalled and actual baseline scores from the same subjects might have been caused by some systematic mechanism.

	RESULTS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION APPENDIX: UCLA Prostate Cancer... REFERENCES

 
Of the 123 patients contacted for this study, 107 returned a survey, for a response rate of 87%. Mean age of the respondents was 61.5 years, and mean time since surgery was 21.1 months (range, 7 to 37 months). Most patients were white, highly educated, and affluent. Respondents were less likely than nonrespondents to hold a college degree (64% v 93%, P = .02), but the groups did not differ in age, race, income, marital or employment status, or time to follow-up (Table 1).

Table 2 lists the internal consistency reliabilities for the five short-form scales as calculated from the baseline and recall surveys. Correlations and internal consistencies were uniformly high, all exceeding the generally accepted standard of 0.70.¹⁹ In this sample, the proportion of variation in the SF-36 component scales accounted for by the SF-12 was 0.93 for the physical component and 0.92 for the mental component. The proportion of variation in the original PCI scales accounted for by each of the corresponding short-form versions was 95% to 96% for all three domains (urinary, bowel, and sexual function).

In all domains, patients recalled their baseline HRQOL as better than it actually was (Table 3). The magnitudes of difference between recalled and actual baseline HRQOL were most striking for sexual and urinary function. Patients remembered their erections and continence as better than they were at baseline. The correlations between recalled and actual baseline HRQOL were modest, but none of them exceeded 0.60. Spearman rank correlations between recalled and actual baseline HRQOL exhibited similar results (data not shown). This suggests the relatively weak concordance between modes of assessment was not dependent on scaling assumptions because the relative rankings of scores were not strongly consistent between the two methods.

Table 4 lists regression coefficients for the effect of age, time since surgery, and education on recall of the various HRQOL domains. Because recall tends to overestimate actual baseline, negative coefficients imply that there is less discrepancy between recall and actual baseline. For example, patients with more education recalled more accurately in all domains. Younger men also recalled more accurately but only on the mental component summary.

Table 5 lists the regression models for prediction of baseline HRQOL scores. Model 1 regresses the actual baseline scale on recall only; models 2 through 4 include recall plus age, time since baseline, or education, respectively; model 5 includes all four predictors simultaneously. Between 17% and 33% of the variation in baseline HRQOL was accounted for by recall alone (model 1). The addition of education brought the range of R² to 21% to 38% (model 4). No systematic differences were seen by age or time since surgery over the 6- to 37-month recall interval. That is, after controlling for other factors, younger patients did not recall their baseline HRQOL any better than older patients. Likewise, beyond 3 months, patients who underwent surgery more recently were equally as poor at recalling actual baseline HRQOL as patients who underwent surgery in the more distant past.

We conducted a sensitivity analysis to address the observations that nonrespondents were more likely to be college educated, and college-educated respondents in our sample had more accurate recall. We imputed recall scores for our nonrespondents based on their age, education, and baseline scores and then incorporated these scores into the models. The difference in R² values for model 5 ranged from a 0.03 decrease to a 0.04 increase, not appreciably altering our results.

	DISCUSSION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION APPENDIX: UCLA Prostate Cancer... REFERENCES

 
Our study demonstrates that, in men undergoing radical prostatectomy for early-stage prostate cancer, recall of pretreatment quality of life is generally poor when compared with measurements collected at the time of surgery. Between 17% and 33% of the variance in actual baseline HRQOL was explained by recalled baseline HRQOL. Better-educated patients seemed to recall somewhat better, but age and time since surgery made no difference.

When studying men with prostate cancer for the effects of surgery on HRQOL, prospective documentation of HRQOL domains is best, but, if healthy control groups are unavailable and retrospective assessment is the only option, there seems to be no particular advantage in surveying patients sooner rather than later during the follow-up period. Nonetheless, others have shown that with increasing time since baseline, recall bias is more likely to be driven by personality traits than the actual health state in question.²⁰

Our findings are consistent with those of Aseltine et al,²¹ who examined the effect of recall among patients treated for benign prostatic hyperplasia (BPH). They compared perceived changes in health since surgical treatment with actual before/after comparisons. Their analysis showed that patients' retrospective reports of change in overall health were more favorable than actual before/after measurements indicated. Sixty-two percent reported that their health had improved 3 months after surgery, although only 33% of the same patients' actual prospective measurements indicated increases in self-rated health. Yet, retrospective changes in how patients felt showed good agreement with prospective symptom assessment; 72% of patients reported feeling better 3 months after surgery, and 84% of these patients showed improvements in a prospectively measured symptom index. An important limitation of the BPH work was that the authors used different measures to assess the accuracy of recall. Rather than comparing recall reporting with actual baseline for the same measures, a global assessment of perceived change was compared with changes in health ratings and symptoms scores. In a similar study, Emberton et al²² used the American Urological Association symptom index²³ in a prospective study of 61 men treated surgically for BPH. Correlations between actual and recalled voiding symptoms demonstrated only modest agreement (r = .60) before and after transurethral prostatectomy.

The problem of recall bias is known to impact studies of sexual function and behavior. Correlations between self- and partner-reported sexual activity decrease significantly over time.²⁴ It may also affect studies involving recollection of screening behaviors for prostate cancer²⁵ and cervical cancer,²⁶ exposure to second-hand smoke,²⁷ and family cardiac risk factors.²⁸ This effect, sometimes termed "hindsight bias,"²⁹ is based on deeply ingrained patterns in memory function and may significantly affect the quality of studies that rely on recalled information.^30,31 Other studies suggest that, for some factors, recall bias may not have as great an impact. In particular, recall of sociodemographic variables³² may be more stable over periods as long as 50 years.³³

The underlying explanation for our findings may be related to what has recently been called response shift. That is, as time progresses and patients adapt to new impairments, they may experience an altered frame of reference that distorts recall. Thus, the chronic nature of these impairments may lead patients to take on a romanticized perspective of their baseline quality of life.³⁴

Our study has several important limitations. First, general application of our conclusions may be hampered by the fact that our patients were predominantly white, highly educated, and affluent. However, if this demographic profile were hypothesized to improve recall, the weight of our findings would be underscored. Second, we studied only men who had undergone surgery for early-stage tumors. Our results may have been different in men undergoing radiotherapy, undergoing watchful waiting for their prostate cancer, or with advanced disease. In addition, despite controlling for age, education, and time since surgery, our analysis might have been affected by other unmeasured biases.

The most methodologically rigorous way of reporting quality-of-life outcomes in men treated for early-stage prostate cancer is to use patients as their own controls. By collecting data before treatment and observing subjects longitudinally, investigators can ensure that HRQOL changes are analyzed in the context of any impairment that might have been present at baseline. If baseline assessment is not feasible, then asking patients to recall their pretreatment HRQOL must be undertaken with the caveat that there may be systematic bias and substantial random error. Because patients tend to see the past through rose-colored glasses, this recall bias must be adjusted or addressed. Alternatively, a control group of men who are similar in age and other demographic variables may be used for comparison.

	APPENDIX: UCLA Prostate Cancer Index—Short Form

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION APPENDIX: UCLA Prostate Cancer... REFERENCES

 
Urinary Function

1. Over the past 4 weeks, how often have you leaked urine?
    Every day; About once a week; Less than once a week; Not at all

2. Which of the following best describes your urinary control during the last 4 weeks?
    No control whatsoever; Frequent dribbling; Occasional dribbling; Total control

3. How many pads or adult diapers per day did you usually use to control leakage during the last 4 weeks?
    3 or more pads per day; 1-2 pads per day; No pads

4. How big a problem, if any, has dripping urine or wetting your pants been for you during the last 4 weeks?
    No problem; Very small problem; Small problem; Moderate problem; Big problem

Urinary Bother

1. Overall, how big a problem has your urinary function been for you during the last 4 weeks?
    No problem; Very small problem; Small problem; Moderate problem; Big problem

Bowel Function

1. How often have you had rectal urgency (felt like you had to pass stool, but did not) during the last 4 weeks?
    More than once a day; About once a day; More than once a week; About once a week; Rarely or never

2. How much distress have your bowel movements caused you during the last 4 weeks?
    Severe distress; Moderate distress; Little distress; No distress

3. How often have you had crampy pain in your abdomen or pelvis during the last 4 weeks?
    Several times a day; About once a day; Several times a week; About once a week; About once this month; Rarely or never

Bowel Bother

1. Overall, how big a problem have your bowel habits been for you during the last 4 weeks?
    No problem; Very small problem; Small problem; Moderate problem; Big problem

Sexual Function

How would you rate each of the following during the last 4 weeks?

1. Your ability to have an erection?
    Very poor; Poor; Fair; Good; Very Good

2. Your ability to reach orgasm (climax)?
    Very poor; Poor; Fair; Good; Very Good

3. How would you describe the usual QUALITY of your erections?
    None at all; Not firm enough for any sexual activity; Firm enough for masturbation and foreplay only; Firm enough for intercourse

4. How would you describe the FREQUENCY of your erections?
    I NEVER had an erection when I wanted one;
    I had an erection LESS THAN HALF the time I wanted one;
    I had an erection ABOUT HALF the time I wanted one;
    I had an erection MORE THAN HALF the time I wanted one;
    I had an erection WHENEVER I wanted one

5. Overall, how would you rate your ability to function sexually during the last 4 weeks?
    Very poor; Poor; Fair; Good; Very good

Sexual Bother

1. Overall, how big a problem has getting and maintaining an erection been for you during the last 4 weeks?
    No problem; Very small problem; Small problem; Moderate problem; Big problem

	ACKNOWLEDGMENTS

 
We thank Sally Carson for statistical support and the many patients who selflessly gave their time and energy to participate in this study.

	NOTES

 
Funded by awards from the University of California Cancer Research Coordinating Committee and the Bing Fund.

	REFERENCES

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION APPENDIX: UCLA Prostate Cancer... REFERENCES

 
1. Landis SH, Murray T, Bolden S, et al: Cancer statistics, 1998. CA Cancer J Clin 48:6-29, 1998 (published erratum appears in CA Cancer J Clin 48:192, 1998) [Abstract]

2. Feldman HA, Goldstein I, Hatzichristou DG, et al: Impotence and its medical and psychosocial correlates: Results of the Massachusetts Male Aging Study. J Urol 151:54-61, 1994[Medline]

3. Thom D: Variation in estimates of urinary incontinence prevalence in the community: Effects of differences in definition, population characteristics, and study type. J Am Geriatr Soc 46:473-480, 1998[Medline]

4. Mulligan T, Moss CR: Sexuality and aging in male veterans: A cross-sectional study of interest, ability, and activity. Arch Sex Behav 20:17-25, 1991[Medline]

5. Helgason AR, Adolfsson J, Dickman P, et al: Sexual desire, erection, orgasm and ejaculatory functions and their importance to elderly Swedish men: A population-based study. Age Ageing 25:285-91, 1996[Abstract/Free Full Text]

6. O'Keefe EA, Talley NJ, Zinsmeister AR, et al: Bowel disorders impair functional status and quality of life in the elderly: A population-based study. J Gerontol A Biol Sci Med Sci 50:M184-M189, 1995

7. Kay L, Jorgensen T, Jensen KH: The epidemiology of irritable bowel syndrome in a random population: Prevalence, incidence, natural history and risk factors. J Intern Med 236:23-30, 1994[Medline]

8. Litwin MS, Hays RD, Fink A, et al: Quality-of-life outcomes in men treated for localized prostate cancer. JAMA 273:129-135, 1995[Abstract]

9. Herrmann D: Reporting current, past, and changed health status: What we know about distortion. Med Care 33:AS89-AS94, 1995

10. Litwin MS, McGuigan KA, Shpall AI, et al: Recovery of health related quality of life in the year after radical prostatectomy: Early experience. J Urol 161:515-519, 1999[Medline]

11. Hays RD, Sherbourne CD, Mazel RM: The RAND 36-Item Health Survey 1.0. Health Econ 2:217-227, 1993[Medline]

12. Ware JE Jr, Sherbourne CD: The MOS 36-item short-form health survey (SF-36): I. Conceptual framework and item selection. Med Care 30:473-483, 1992[Medline]

13. Stewart AL, Hays RD, Ware JE Jr: The MOS short-form general health survey: Reliability and validity in a patient population. Med Care 26:724-735, 1988[Medline]

14. Litwin MS, Hays RD, Fink A, et al: The UCLA Prostate Cancer Index: Development, reliability, and validity of a health-related quality of life measure. Med Care 36:1002-1012, 1998[Medline]

15. Lubeck DP, Litwin MS, Henning JM, et al: Measurement of health-related quality of life in men with prostate cancer: The CaPSURE database. Qual Life Res 6:385-392, 1997[Medline]

16. Ware J Jr, Kosinski M, Keller SD: A 12-Item Short-Form Health Survey: Construction of scales and preliminary tests of reliability and validity. Med Care 34:220-33, 1996[Medline]

17. Ware JE, Kosinski M, Keller SD: SF-12: How to Score the SF-12 Physical and Mental Health Summary Scales. Boston, MA, The Health Institute, New England Medical Center, 1995

18. Cronbach LJ: Coefficient alpha and the internal structure of tests. Psychometrika 16:297-334, 1951

19. Nunnally JC: Psychometric Theory (ed 2). New York, NY, McGraw-Hill, 1978

20. Huisman SJ, van Damm FSAM, Aaronson NK, et al: On measuring complaints of cancer patients: Some remarks on the time span of the question, in Aaronson NK, Beckman J (eds): The Quality of Life in Cancer Patients. New York, NY, Raven Press, 1987, pp 101-109

21. Aseltine RH Jr, Carlson KJ, Fowler FJ Jr, et al: Comparing prospective and retrospective measures of treatment outcomes. Med Care 33:AS67-AS76, 1995

22. Emberton M, Challands A, Styles RA, et al: Recollected versus contemporary patient reports of pre-operative symptoms in men undergoing transurethral prostatic resection for benign disease. J Clin Epidemiol 48:749-756, 1995[Medline]

23. Barry MJ, Fowler FJ Jr, O'Leary MP, et al: The American Urological Association symptom index for benign prostatic hyperplasia: The Measurement Committee of the American Urological Association. J Urol 148:1549-1557, 1992[Medline]

24. Ellish NJ, Weisman CS, Celentano D, et al: Reliability of partner reports of sexual history in a heterosexual population at a sexually transmitted diseases clinic. Sex Transm Dis 23:446-452, 1996[Medline]

25. Ward JE, Hughes AM, Hirst GH, et al: Men's estimates of prostate cancer risk and self-reported rates of screening. Med J Aust 167:250-253, 1997[Medline]

26. Ward J, Sanson-Fisher R: Accuracy of women's recall of opportunistic recruitment for cervical cancer screening in general practice. Aust N Z J Public Health 21:335-336, 1997[Medline]

27. Lee PN: Difficulties in assessing the relationship between passive smoking and lung cancer. Stat Methods Med Res 7:137-163, 1998[Abstract/Free Full Text]

28. Silberberg JS, Wlodarczyk J, Fryer J, et al: Correction for biases in a population-based study of family history and coronary heart disease: The Newcastle Family History Study I. Am J Epidemiol 147:1123-1132, 1998[Abstract/Free Full Text]

29. Pohl RF, Gawlik B: Hindsight bias and the misinformation effect: Separating blended recollections from other recollection types. Memory 3:21-55, 1995[Medline]

30. Chouinard E, Walter S: Recall bias in case-control studies: An empirical analysis and theoretical framework. J Clin Epidemiol 48:245-254, 1995[Medline]

31. Barry D: Differential recall bias and spurious associations in case/control studies. Stat Med 15:2603-2616, 1996[Medline]

32. Krieger N, Okamoto A, Selby JV: Adult female twins' recall of childhood social class and father's education: A validation study for public health research. Am J Epidemiol 147:704-708, 1998[Abstract/Free Full Text]

33. Berney LR, Blane DB: Collecting retrospective data: Accuracy of recall after 50 years judged against historical records. Soc Sci Med 45:1519-1525, 1997

34. Breetvelt IS, Van Dam FS: Underreporting by cancer patients: The case of response-shift. Soc Sci Med 32:981-987, 1991

Submitted November 30, 1998; accepted May 10, 1999.

This article has been cited by other articles:

				J. C. Hu In Reply J. Clin. Oncol., October 20, 2008; 26(30): 5001 - 5002. [Full Text] [PDF]

				E. Basch, D. Artz, D. Dulko, K. Scher, P. Sabbatini, M. Hensley, N. Mitra, J. Speakman, M. McCabe, and D. Schrag Patient Online Self-Reporting of Toxicity Symptoms During Chemotherapy J. Clin. Oncol., May 20, 2005; 23(15): 3552 - 3561. [Abstract] [Full Text] [PDF]

				D. C. Miller, M. G. Sanda, R. L. Dunn, J. E. Montie, H. Pimentel, H. M. Sandler, W. P. McLaughlin, and J. T. Wei Long-Term Outcomes Among Localized Prostate Cancer Survivors: Health-Related Quality-of-Life Changes After Radical Prostatectomy, External Radiation, and Brachytherapy J. Clin. Oncol., April 20, 2005; 23(12): 2772 - 2780. [Abstract] [Full Text] [PDF]

				J. A. Talcott, J. Manola, J. A. Clark, I. Kaplan, C. J. Beard, S. P. Mitchell, R. C. Chen, M. P. O'Leary, P. W. Kantoff, and A. V. D'Amico Time Course and Predictors of Symptoms After Primary Prostate Cancer Therapy J. Clin. Oncol., November 1, 2003; 21(21): 3979 - 3986. [Abstract] [Full Text] [PDF]

				A. S. Hamilton, J. L. Stanford, F. D. Gilliland, P. C. Albertsen, R. A. Stephenson, R. M. Hoffman, J. W. Eley, L. C. Harlan, and A. L. Potosky Health Outcomes After External-Beam Radiation Therapy for Clinically Localized Prostate Cancer: Results From the Prostate Cancer Outcomes Study J. Clin. Oncol., May 1, 2001; 19(9): 2517 - 2526. [Abstract] [Full Text] [PDF]

				B. L. Dalkin Re: Health Outcomes After Prostatectomy or Radiotherapy for Prostate Cancer: Results From the Prostate Cancer Outcomes Study J Natl Cancer Inst, March 7, 2001; 93(5): 401 - 401. [Full Text] [PDF]

				A. L. Potosky, J. Legler, R. M. Hoffman, and F. D. Gilliland RESPONSE: Re: Health Outcomes After Prostatectomy or Radiotherapy for Prostate Cancer: Results From the Prostate Cancer Outcomes Study J Natl Cancer Inst, March 7, 2001; 93(5): 401a - 402a. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Litwin, M. S. Articles by McGuigan, K. A. Search for Related Content PubMed PubMed Citation Articles by Litwin, M. S. Articles by McGuigan, K. A.