HomeLarge Type Edition
HOME ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Services
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
PubMed
Right arrow PubMed Citation
The Journals of Gerontology Series A: Biological Sciences and Medical Sciences 57:M100-M105 (2002)
© 2002 The Gerontological Society of America

Seniors' Preferences for Cancer Screening and Medication Use Based on Absolute Risk Reduction

Donald J. Murphya, Gregory J. Gahma, Sara Santillib, Patricia Northc, Scott C.N. Oliverd and Howard Shapiroe

a Presbyterian/St. Luke's Medical Center and GeriMed of America, Inc., Denver, Colorado
b St. Joseph Healthcare, Albuquerque, New Mexico
c Hilton Head Medical Center, Hilton Head, South Carolina
d University of Colorado Health Sciences Center, Denver
e HealthONE Alliance Research, Denver, Colorado

Donald J. Murphy, 1721 E. 19th Ave., Suite 574, Denver, CO 80218 E-mail: DonJMurphy{at}aol.com.


   Abstract
Top
 Abstract
Methods
 Results
 Discussion
 References
 
Background. This study was conducted to determine the influence of patient perceptions of absolute risk on choices for cancer screening and use of medications to prevent heart attack, stroke, and hip fracture.

Methods. At the end of routine office visits, we surveyed all eligible consecutive patients who visited four geriatricians in a Denver practice between November 8, 1993, and February 9, 1994.

Results. We saw a total of 675 outpatients during the study period and completed the interview with 409 patients (75% female, mean age 81, 78% Caucasian). We found a strong correlation between (i) increased probability of detecting cancer and greater preference for cancer screening tests (p < .001) and (ii) increased probability of preventing disease (heart attack, stroke, or hip fracture) and greater preference for preventive medication (p < .0001). There was notable variability in seniors' preferences for a given therapy at each absolute risk threshold. For example, 15% of seniors did not think that highly effective, inexpensive medications to prevent heart attacks were worthwhile for them. At the other end of the spectrum, 22% of seniors felt that low-yield, costly medications to prevent heart attacks were worthwhile.

Conclusions. Seniors readily understand the probability of benefit expressed in terms of absolute risk reduction. Furthermore, probability of benefit strongly influences seniors' preferences for cancer screening and preventive medication use. Finally, there is variety in the thresholds of prevention at which individuals are willing to accept preventive treatment. The probability of benefit is an essential and useful element for seniors to make informed decisions about routine health services.

HEALTH care providers routinely and frequently screen patients for cancer and prescribe medications to prevent disease. Many guidelines are available to shape recommendations for detecting early disease and for preventing the progression of disease (1)(2)(3)(4). However, ethical justification for preventive interventions, such as cancer screening and preventive medication, depends not only on the benefits to the sick, but also on risks and costs to the healthy (5). Surprisingly few data are available that indicate how patients value the risk-reduction benefit of preventive interventions.

Respect for patient autonomy has emerged as a guiding principle in medical practice. A tangible manifestation of this principle is informed consent, which recent publications have emphasized as an important component in selecting preventive interventions (6)(7)(8)(9)(10). In 1989, Brett distinguished the effects of statistical reporting methods, asserting that "physicians and patients are likely to perceive a relative benefit of 20 or 30 percent more favorably than an absolute benefit of 1 or 2 percentage points" (11). To meaningfully and adequately inform a patient about the individual benefit of a preventive intervention, the practitioner must present the absolute risk reduction (12)(13)(14)(15).

As primary care providers, we wanted to determine the influence of absolute risk reduction information on our patients' preferences for cancer screening and the use of medications to prevent common diseases, such as myocardial infarction, stroke, and hip fracture. The impact of the probability of benefit on these preferences is important for several reasons. First, a correlation between increased absolute risk reduction and greater preference for a preventive intervention reinforces the value of including statistical information in helping a patient make an informed decision. Second, variability in response to prevention odds reveals that certain patients may or may not follow the trend, thus emphasizing the importance of individualized care. A decision to treat will depend in part on the patient's value of risk reduction and not simply on a universal recommendation based only on statistical evidence. This individualized approach may elucidate problems with patient care. For example, noncompliance may be an obstacle if a patient values risk reduction differently than a group of experts or a primary care physician. Finally, more information about patients' preferences regarding preventive interventions should help shape future guidelines, particularly as new preventive therapies arise (16).


   Methods
Top
 Abstract
 Methods
Results
 Discussion
 References
 
We designed a questionnaire based on discussions with patients followed by four clinicians in a hospital-based geriatrics clinic in downtown Denver. We finalized the questionnaire after a pilot test on 30 other patients from the geriatrics practice.

We surveyed all eligible consecutive patients who visited three clinicians between November 8, 1993, and February 9, 1994. The fourth clinician had a nurse practitioner interview all of his eligible consecutive patients at a satellite clinic. Interviewers included three physicians, one geriatric nurse practitioner, and two medical residents. We conducted the interviews at the end of routine office visits. Most interviews lasted 5 to 8 minutes.

The questionnaire was divided into five sections. The first section focused on the patient's preferences for cancer screening that involved (i) a simple blood test; (ii) a test requiring an extra appointment and some discomfort; and (iii) a similar test requiring out-of-pocket expenses (i.e., ones that "put a strain on the pocketbook that month").

The second, third, and fourth sections focused on the patient's preferences for medications to prevent heart attacks, strokes, and hip fractures, respectively. In each of these sections, we presented one question when the patient assumed the medication costs were very small. We presented another question when the patient assumed the cost of the medication "put a strain on the pocketbook." We used the same thresholds for absolute risk reduction (50%, 10%, 5%, 2%, 1%, and 0.02%) in each of these three sections. We presented these thresholds in terms of probabilities, e.g., "a 1 in 2 chance that the medicine would prevent a heart attack in you" or "a 1 in 100 chance that the medicine would prevent a heart attack in you."

At the beginning of each of the first four sections, we asked the patients how worried they were about getting cancer, having a heart attack, having a stroke, or having a hip fracture, respectively. We recorded their responses on a 5-point Likert scale.

In the fifth section, we asked demographic information, determined whether they were aware of having certain risk factors for cardiovascular disease (e.g., smoking, diabetes, cholesterol problems), and asked two general questions about their perceived quality of health and quality of life.

To test the reproducibility of the questionnaire, we repeated the interview for 30 patients 2 months after the initial interview. We considered responses concordant if they were the same in each interview or if the responses differed by one threshold in either direction. For example, a patient whose threshold was 1 in 50 in the first interview had a concordant response if his or her threshold was 1 in 20, 1 in 50, or 1 in 100 in the second interview. The concordance rate for all of the questions regarding thresholds was 80%.

We used the SAS software system (17) to analyze the data. We analyzed continuous variables with Student's t test and categorical variables with the chi-square test. We used a linear trend for proportions to analyze the cancer screening preferences. Because each patient responded for each test complexity, we used a matched design according to the methods of Fleiss (18). We analyzed the medication preferences by McNemar's test for paired dichotomous data (18).

Within each of the hypothetical medical conditions, we addressed the multiple testing issue through use of the Bonferroni adjustment (19) based on the number of comparisons made within that condition (four for screening and six for prevention). For cancer screening preferences and medication preferences, statistical significance was assumed for p < .0125 and p < .0083, respectively.


   Results
Top
 Abstract
 Methods
 Results
Discussion
 References
 
The participating clinicians saw a total of 675 outpatients during the period of study. We excluded 266 patients for the following reasons: participation in the pilot study (n = 22); dementia (106); inability to fully comprehend the questions (30); lack of time to complete the interview (50); and miscellaneous (language barrier, acute illness, or mental illness) (58). We completed the interview with 409 patients (75% female, mean age 81, 78% Caucasian).

Fig. 1 illustrates their thresholds for cancer screening and shows a clear association between likelihood of detection and preference for screening (p < .001). The percentage of seniors opting for a screening test increased as the likelihood of detection increased. As expected, the complexity and cost of the screening test influence the threshold at which seniors thought screening was worthwhile. Of note is the percentage of seniors at either extreme of the scale. Some seniors (13%) would want cancer screening even if the likelihood of detection was remote (i.e., 1 in 10,000). On the other hand, some seniors (18%) would not want a screening test even if the likelihood of detection was high (i.e., 1 in 10).



View larger version (23K):
[in this window]
[in a new window]
  		Figure 1. Percentage of senior patients (N = 409) accepting a blood test, complex test, and test costly to the patient at various odds of detecting cancer.

 
Fig. 2 illustrates our patients' thresholds for preferring medications that prevent heart attacks, strokes, and hip fractures. Again, there is a clear association between the likelihood of prevention and the preference for medications (p < .0001). The percentage of seniors opting to take medications to prevent diseases increases as the absolute risk reduction increases. The percentage of seniors at either end of the spectrum is noteworthy. For example, 15% of seniors do not think that highly effective, inexpensive medications to prevent heart attacks are worthwhile for them. Alternatively, 22% of seniors think that low-yield, costly medications to prevent heart attacks are worthwhile for them.



View larger version (27K):
[in this window]
[in a new window]
  		Figure 2. Percentage of senior patients (N = 409) accepting inexpensive ($) and expensive ($$$) medications at various odds of preventing heart attack, stroke, and hip fracture.

 
We found no correlation between their concern about a disease state, their awareness of certain cardiovascular risk factors, their perceived quality of health, their perceived quality of life, and their preferences for cancer screening or medications to prevent disease. The only factor that correlated with thresholds for cancer screening and medication use was age. Older patients were less likely to desire cancer screening (p < .001) and medication use (p < .005) than younger patients.


   Discussion
Top
 Abstract
 Methods
 Results
 Discussion
References
 
Our results indicate that seniors' preferences for cancer screening and medications to prevent disease depend on the likelihood that these measures will help them. The percentage opting for these measures increases as the likelihood of benefit (i.e., the odds of detecting cancer or the absolute risk reduction for medications to prevent disease) increases. Some seniors opt for these measures even if the likelihood of benefit is very low and the cost is high. Others, in contrast, do not think these measures are worthwhile for themselves even if the likelihood of benefit is high and the cost negligible.

The presentation of statistical data can significantly influence providers (20)(21)(22)(23)(24) and patients' (15)(25)(26)(27) preferences for treatment. Relative risk reduction, or the degree to which an intervention diminishes the risk of a disease, is insignificant without a comprehension of the baseline incidence risk for that disease. The absolute risk reduction relates the benefit of the intervention to the baseline risk, thereby providing a more realistic idea of the real size of the benefit to a patient (12). Unfortunately, many providers do not discuss risk reduction in quantitative terms that patients understand (28)(29). Quite possibly, many providers do not grasp the difference between absolute and relative risk-reduction and, furthermore, are unfamiliar with the risk-reduction statistics for many screening and preventive medication therapies.

This study indicates that when seniors are presented with absolute risk-reduction information, their treatment preferences may differ from the standardized screening recommendations espoused by professional societies. For example, mammography, which will detect breast cancer in 1 of every 125 older women screened, could be considered complex (i.e., it requires an extra appointment and some discomfort) and costly (30)(31). When offered a complex, costly cancer screening test with a detection rate of 1 in 100, 48–62% of seniors would decline the procedure. We recognize that a patient's response to the probability of detecting cancer may be different than his or her response to the probability of preventing morbidity or mortality. Utilizing the number needed to screen to prevent one premature cancer death, which varies for mammography from 1700 to 68,000, might have provided patients with a better idea of the likelihood that the intervention would benefit them individually (12)(32)(33)(34)(35). Nevertheless, the fact that about half of our respondents would decline cancer screening at a threshold similar to the detection rate for mammography indicates the importance of eliciting patient preferences before proceeding with screening.

Antihypertensives are the most widely prescribed medicines in geriatrics. Treating hypertension with medications has a 1 in 50 to a 1 in 20 chance of preventing strokes in seniors (i.e., a 2–5% absolute risk reduction) (36)(37)(38). When considering inexpensive medications associated with an absolute risk reduction for stroke of 2–5%, 41–51% of our respondents would decline such medications, while 54–65% would decline a costly medication. Similar responses occurred for cholesterol-lowering agents to prevent myocardial infarctions, which have a 1 in 100 to a 1 in 20 chance of preventing myocardial infarction in seniors, depending on premorbid status (39)(40)(41).

The absolute risk reduction associated with estrogen replacement therapy may be 1–2% (42)(43)(44)(45)(46)(47). When seniors consider this likelihood of benefit, 70–78% (depending on cost) would decline.

Medical decision making is certainly more complex than simply considering probabilities, in this case cancer detection rates and absolute risk reductions. Dolan and Bordley have presented a decision-making model that considers the major components of decision making in ambulatory care, including risk reduction, avoiding side effects, minimizing out-of-pocket costs, and avoiding hassles (48). Based on our experience in the pilot study, in which consideration of side effects seemed to unduly bias patients against medication use, we excluded this component from our study. The reality is that treatment of hypertension, for example, rarely needs to involve long-term side effects because we have so many treatment options.

Other factors impacting medical decision making that we did not explore in this study include belief in certain diagnostics (consider, for example, the controversy over mammography for women age 40 to 49 (49)(50)(51)(52)), belief in certain therapeutics (53), emotion (54), burden on family, association with close friends and relatives who have had particular diagnoses, expected quality of life (55), expected symptoms (53), competing risks (56), and cost to society. Our exploration of factors such as fear about a particular disease, awareness of risk factors, perceived quality of life, and perceived quality of health was cursory and did not result in any correlation between these factors and the preventive intervention about which we asked.

Our results from this study and from a previous study to determine preferences for cardiopulmonary resuscitation (57) suggest that probability of benefit is an important factor for seniors to make informed decisions about their medical care. This study, however, has several important limitations.

First, the proposed preventive interventions were hypothetical and nonspecific. Had we solicited patient preferences for distinct, well-known cancer screening tests (e.g., mammography) or preventive medications (e.g., aspirin to prevent myocardial infarction), their responses may have differed. Future inquiry into seniors' preferences for specific preventive interventions with known indications, costs, side effects, and absolute risk reductions would be highly beneficial to the practitioner in terms of anticipating the likelihood that a patient would be willing to accept the intervention.

Second, the survey did not evaluate the actual risk factors or disease states of our patients, nor did we correlate such information with their preference for corresponding preventive interventions. If, for example, we had asked the questions to a control group and a group of patients with hypertension who were already taking antihypertensive medications, the responses between the two groups probably would have differed. The effect of patients' risk factors or disease states on their preference for preventive interventions needs more exploration.

Third, our results came from seniors who were not acutely ill, had well-established relationships with their physicians, and were predominantly of one ethnicity (78% Caucasian). The conclusions may not be generalizable to other populations.

Fourth, we may have created a framing bias with the use of terms such as, "a simple blood test," "a test requiring an extra appointment and some discomfort," and "put a strain on the pocketbook that month." Additionally, answer choices for the survey were limited to a few specific risk-reduction percentages. Rather than spontaneously generating their own threshold for risk reduction for a particular intervention, respondents had to choose from prescribed percentages, which precluded an in-depth linear regression analysis of patient-generated thresholds at which preference for therapy rapidly diminished. Future work could determine if there are threshold probabilities of prevention that lead to optimal interest in the intervention.

Finally, we did not specify the frequency of cancer screening or the duration of medication use in our hypothetical questions. When considering cancer screening, our patients may have assumed one-time screening, or they may have assumed repeated screening over many years. Similarly, when considering medication use, they may have assumed limited duration of treatment (e.g., 1 year), or they may have assumed prolonged duration (e.g., 10 to 20 years). This factor could have contributed to some of the variability in patient responses.

Despite these limitations, we conclude that seniors are responsive to probability of benefit, expressed in absolute numbers, that such probabilities correlate with interest in the intervention, and that there is a diversity of preferences at each probability threshold, which necessitates individualized care. Absolute risk benefit is a key element in medical decision making with seniors.

Our results have encouraged us to learn better ways to communicate benefits and burdens to our patients. Expressing absolute risk reduction with simple percentages has been effective and easy to follow for clinicians and patients. In doing so, we have found significant variation in how seniors value risk reduction. We have also found that the preferences of a majority of our oldest patients are not consistent with many clinical guidelines (e.g., yearly mammograms with no upper age limit) espoused by professional societies. Our study suggests that, if clinical guidelines are to be pertinent and useful to our patients, professionals must obtain patient input in preventive treatment decisions.


   Acknowledgments
 
We are grateful to the staff and patients at the Presbyterian/St. Luke's Senior Health Care Center for their participation; to the University of Colorado Cancer Center and the American Cancer Society for financial support; and to Vikki Jett for help preparing the manuscript.

Received July 24, 2000

Accepted April 16, 2001


   References
Top
 Abstract
 Methods
 Results
 Discussion
 References
 

  1. Woolf SH, Kamerow DB, Lawrence RS, Medalie JH, Estes EH, 1990. The periodic health examination of older adults: the recommendations of the U.S. Preventive Services Task Force. Part II. Screening tests. J Am Geriatr Soc. 38:933-942. [Medline]
  2. Sox HC, Jr 1994. Preventive health services in adults. N Engl J Med. 330:1589-1595. [Free Full Text]
  3. Hayward RSA, Steinberg EP, Ford DE, et al. 1991. Preventive care guidelines: 1991. Ann Intern Med. 114:758-783.
  4. Goldberg TH, Chavin SI, 1997. Preventive medicine and screening in older adults. J Am Geriatr Soc. 45:344-354. [Medline]
  5. Malm HM, 1999. Medical screening and the value of early detection: when unwarranted faith leads to unethical recommendations. Hastings Cent Rep. 29:26-37. [Medline]
  6. Forrow L, Wartman SA, Brock DW, 1988. Science, ethics, and the making of clinical decisions. Implications for risk factor intervention. JAMA. 259:3161-3167. [Abstract/Free Full Text]
  7. O'Hagan J, 1991. The ethics of informed consent in relation to prevention screening programmes. N Z Med J. 104:121-123. [Medline]
  8. Wolf AMD, Nasser JF, Wolf AM, Schorling JB, 1996. The impact of informed consent on patient interest in prostate-specific antigen screening. Arch Intern Med. 156:1333-1336. [Abstract/Free Full Text]
  9. Ubel PA, 1996. Informed consent: from bodily invasion to the seemingly mundane. Arch Intern Med. 156:1262-1263. [Abstract/Free Full Text]
  10. Wolf AMD, Becker DM, 1996. Cancer screening and informed patient discussions: truth and consequences. Arch Intern Med. 156:1069-1072. [Abstract/Free Full Text]
  11. Brett AS, 1989. Treating hypercholesterolemia: how should practicing physicians interpret the published data for patients?. N Engl J Med. 321:676-680. [Medline]
  12. Marshall KG, 1996. Prevention. How much harm? How much benefit? 1. Influence of reporting methods on perception of benefits. Can Med Assoc J. 154:1493-1499. [Abstract]
  13. Laupacis A, 1996. Preventive therapies: weighing the pros and cons. Can Med Assoc J. 154:1510-1512. [Abstract]
  14. Rajkumar SV, Sampathkumar P, Gustafson AB, 1996. Number needed to treat is a simple measure of treatment efficacy for clinicians. J Gen Intern Med. 11:357-359. [Medline]
  15. Schwartz LM, Woloshin S, Black WC, Welch HG, 1997. The role of numeracy in understanding the benefit of screening mammography. Ann Intern Med. 127:966-972. [Abstract/Free Full Text]
  16. Cummings SR, Eckert S, Krueger KA, et al. 1999. The effect of Raloxifene on risk of breast cancer in postmenopausal women: results from the MORE Randomized Trial. JAMA. 281:2189-2197. [Abstract/Free Full Text]
  17. SAS Software System. Version 6.12. Cary, NC: SAS Institute Inc; 1998.
  18. Fleiss JL. Statistical Methods for Rates and Proportions. 2nd ed. New York: Wiley; 1981.
  19. Meinert CL. Clinical Trials. Design, Conduct, and Analysis. Oxford: Oxford University Press; 1986.
  20. Forrow L, Taylor WC, Arnold RM, 1992. Absolutely relative: how research results are summarized can affect treatment decisions. Am J Med. 92:121-124. [Medline]
  21. Hux JE, Levinton CM, Naylor CD, 1994. Prescribing propensity: influence of life-expectancy gains and drug costs. J Gen Intern Med. 9:195-201. [Medline]
  22. Naylor CD, Chen E, Strauss B, 1992. Measured enthusiasm: does the method of reporting trial results alter the perceptions of therapeutic effectiveness?. Ann Intern Med. 117:916-921.
  23. Bucher HC, Weinbacher M, 1994. Influence of method or reporting study results on decision of physicians to prescribe drugs to lower cholesterol concentration. BMJ. 309:761-764. [Abstract/Free Full Text]
  24. Friedmann PD, Brett AS, Mayo-Smith MF, 1996. Differences in generalists' and cardiologists' perceptions of cardiovascular risk and the outcomes of preventive therapy in cardiovascular disease. Ann Intern Med. 124:414-421. [Abstract/Free Full Text]
  25. Malenka DJ, Baron JA, Johansen S, et al. 1993. The framing effect of relative and absolute risk. J Gen Intern Med. 8:543-548. [Medline]
  26. McNeil BJ, Pauker SG, Sox HC, Jr. Tversky A, 1982. On the elicitation of preferences for alternative therapies. N Engl J Med. 306:1259-1262. [Abstract]
  27. Man-Son-Hing M, Laupacis A, O'Connor A, et al. 1996. Warfarin for atrial fibrillation: the patient's perspective. Arch Intern Med. 156:1841-1848. [Abstract/Free Full Text]
  28. Kalet A, Roberts JC, Fletcher R, 1994. How do physicians talk with their patients about risks?. J Gen Intern Med. 9:402-404. [Medline]
  29. Slaytor EK, Ward JE, 1998. How risks of breast cancer and benefits of screening are communicated to women: analysis of 58 pamphlets. BMJ. 317:263-264. [Free Full Text]
  30. Hurley SF, Kaldor JM, 1992. The benefits and risks of mammographic screening for breast cancer. Epidemiol Rev. 14:101[Free Full Text]
  31. Kerlikowske K, Grady D, Barclay J, et al. 1993. Positive predictive value of screening mammography by age and family history of breast cancer. JAMA. 270:2444-2450. [Abstract/Free Full Text]
  32. Skrabanek P, 1989. Mass mammography. The time for reappraisal. Int J Technol Assess Health Care. 5:423-430. [Medline]
  33. Rembold CM, 1998. Number needed to screen: development of a statistic for disease screening. BMJ. 317:307-312. [Abstract/Free Full Text]
  34. Steiner JF, 1999. Talking about treatment: the language of populations and the language of individuals. Ann Intern Med. 130:618-622. [Free Full Text]
  35. Goodman SN, 1999. Probability at the bedside: the knowing of chances or the chances of knowing?. Ann Intern Med. 130:604-606. [Free Full Text]
  36. SHEP Cooperative Research Group1991. Prevention of stroke by antihypertensive drug treatment in older persons with isolated systolic hypertension: final results of the systolic hypertension in the elderly program (SHEP). JAMA. 265:3255-3264. [Abstract/Free Full Text]
  37. Dahlof B, Lindholm LH, Hansson L, et al. 1991. Morbidity and mortality in the Swedish Trial in Old Patients with Hypertension (STOP-Hypertension). Lancet. 338:1281-1285. [Medline]
  38. Working Party MRC, 1992. Medical Research Council trial of treatment of hypertension in older adults: principal results. BMJ. 304:405-412.
  39. Krumholz HM, Seeman TE, Merrill SS, et al. 1994. Lack of association between cholesterol and coronary heart disease mortality and morbidity and all-cause mortality in persons older than 70 years. JAMA. 272:1335-1340. [Abstract/Free Full Text]
  40. Denke MA, Grundy SM, 1990. Hypercholesterolemia in elderly persons: resolving the treatment dilemma. Ann Intern Med. 112:780-792.
  41. Criqui MH, 1991. Cholesterol, primary and secondary prevention, and all-cause mortality. Ann Intern Med. 115:973-976.
  42. Felson DT, Zhang Y, Hannan MT, et al. 1993. The effect of postmenopausal estrogen therapy on bone density in elderly women. N Engl J Med. 329:1141-1146. [Abstract/Free Full Text]
  43. Kiel DP, Felson DT, Anderson JJ, et al. 1987. Hip fracture and the use of estrogens in postmenopausal women: The Framingham Study. N Engl J Med. 317:1169-1174. [Abstract]
  44. Weiss NS, Ure CL, Ballard JH, et al. 1980. Decreased risk of fractures of the hip and lower forearm with postmenopausal use of estrogens. N Engl J Med. 303:1195-1198. [Abstract]
  45. Ettinger B, Genant HK, Cann CE, 1985. Long-term estrogen replacement therapy prevents bone loss and fractures. Ann Intern Med. 102:319-324.
  46. Paganini-Hill A, Ross RK, Gerkins VR, et al. 1981. Menopausal estrogen therapy and hip fractures. Ann Intern Med. 95:28-31.
  47. Naessen T, Persson I, Adami H, et al. 1990. Hormone replacement therapy and the risk for first hip fracture: a prospective, population-based cohort study. Ann Intern Med. 113:95-103.
  48. Dolan JG, Bordley DR, 1993. Involving patients in complex decisions about their care: an approach using the analytic hierarchy process. J Gen Intern Med. 8:204-209. [Medline]
  49. Fletcher SW, 1997. Whither scientific deliberation in health policy recommendations? Alice in the Wonderland of breast-cancer screening. N Engl J Med. 336:1180-1183. [Free Full Text]
  50. Woolf SH, Lawrence RS, 1997. Preserving scientific debate and patient choice: lessons from the Consensus Panel on Mammography Screening. JAMA. 278:2105-2108. [Abstract/Free Full Text]
  51. Ransohoff DF, Harris RP, 1997. Lessons from the mammography screening controversy: can we improve the debate?. Ann Intern Med. 127:1029-1034. [Abstract/Free Full Text]
  52. Phillips K, Glendon G, Knight JA, 1999. Putting the risk of breast cancer in perspective. N Engl J Med. 340:141-144. [Free Full Text]
  53. Mazur DJ, Merz JF, 1996. How older patients' treatment preferences are influenced by disclosures about therapeutic uncertainty: surgery versus expectant management for localized prostate cancer. J Am Geriatr Soc. 44:934-937. [Medline]
  54. Redelmeier DA, Rozin P, Kahneman D, 1993. Understanding patients' decisions: cognitive and emotional perspectives. JAMA. 270:72-76. [Abstract/Free Full Text]
  55. Gage BF, Cardinalli AB, Owens DK, 1996. The effect of stroke and stroke prophylaxis with aspirin or warfarin on quality of life. Arch Intern Med. 156:1829-1836. [Abstract/Free Full Text]
  56. Welch HG, Albertsen PC, Nease RF, Bubolz TA, Wasson JH, 1996. Estimating treatment benefits for the elderly: the effect of competing risks. Ann Intern Med. 124:577-584. [Abstract/Free Full Text]
  57. Murphy DJ, Burrows D, Santilli S, et al. 1994. The influence of the probability of survival on patients' preferences regarding cardiopulmonary resuscitation. N Engl J Med. 330:545-549. [Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 AM J ALZHEIMERS DIS OTHER DEMENHome page
S. J. Cutler and L. G. Hodgson
To test or not to test: Interest in genetic testing for Alzheimer's disease among middle-aged adults
American Journal of Alzheimer's Disease and Other Dementias, January 1, 2003; 18(1): 9 - 20.
[Abstract] [PDF]

Home page
 Journals of Gerontology Series A: Biological Sciences and Medical SciencesHome page
J. E. Morley
Editorial: Hot Topics in Geriatrics
J. Gerontol. A Biol. Sci. Med. Sci., January 1, 2003; 58(1): M30 - 36.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Services
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
PubMed
Right arrow PubMed Citation

HOME ARCHIVE SEARCH TABLE OF CONTENTS