Screening for Prostate Cancer

Clinical Question

Does screening for prostate cancer reduce all-cause or prostate-cancer–specific mortality or impact quality of life?

Evidence-Based Answer

There is insufficient evidence to determine whether screening for prostate cancer reduces mortality or impacts quality of life.

Practice Pointers

This Cochrane review identified two randomized controlled trials (RCTs) comparing mass screening for prostate cancer to no screening: the Quebec trial and the Norrkoping trial. Both studies reported data on prostate cancer incidence, disease-specific mortality, stage at diagnosis, and treatment follow-up. Neither study reported data on quality of life, all-cause mortality, costs, or adverse effects of screening (e.g., false-negative or false-positive results, adverse effects of biopsy, treatment complications). These studies had serious methodologic flaws resulting in a high risk of bias.

The Quebec trial recruited men 45 to 80 years of age from electoral registrations in Quebec City, Canada. A total of 31,133 men were randomly selected to be offered screening, and 15,353 were not invited to screening. Men with prostate cancer and those previously screened and referred to the study's clinic were excluded. Screening was performed annually by a digital rectal examination (DRE) and prostate-specific antigen (PSA) test; by DRE, PSA, and transrectal ultrasound-guided biopsy; or by PSA alone.

The reported relative risk (RR) of death from prostate cancer in men who were screened was 0.39 (95% confidence interval, 0.19 to 0.65). However, only 23 percent of those invited for screening were screened, and 7.3 percent in the nonscreening control group were screened. To address this high rate of crossover from the invited group to the uninvited group, the Cochrane reviewers performed intention-to-treat analysis from the study's raw data. Reanalysis found no difference in prostate-cancer–specific mortality between the two groups at 11-year follow-up.

The second trial recruited men 50 to 69 years of age from Norrkoping, Sweden, from the national population register. A total of 9,026 men were identified, with one in six invited to screening (1,494 invited; 7,532 not invited). Exclusion criteria were not reported. Screening was performed every three years by DRE alone or by DRE and PSA testing. Of those invited for screening, 70 to 77 percent were screened. There were no data on how many uninvited men received screening. In the intention-to-treat analysis, there was no difference between the two groups in prostate-cancer–specific death at 15-year follow-up.

There are two ongoing large-scale RCTs of prostate cancer screening: the European Randomised study of Screening for Prostate Cancer trial and the Prostate, Lung, Colorectal and Ovarian cancer screening trial. It is anticipated that findings from these two trials will be available in the next few years and will provide the evidence needed to determine whether screening for prostate cancer should be offered routinely. In the meantime, shared decision making—including a discussion of the false-positive and false-negative results of screening tests, the variable natural history of prostate cancer in various individuals, and the uncertainty of the net benefit of early detection and treatment in asymptomatic men—has been recommended by major professional organizations.1–3

Vaccines for Preventing Influenza in Older Patients

Clinical Question

Are influenza vaccines safe and effective in persons 65 years and older?

Evidence-Based Answer

According to data obtained primarily from poor-quality observational studies, influenza vaccination in older persons living in long-term care facilities appears to prevent approximately 45 percent of pneumonia cases, hospital admissions, and influenza-related deaths. In older persons living in the community, influenza vaccination prevents about 25 percent of hospitalizations from influenza or respiratory illness. The vaccines appear to be safe.

Practice Pointers

The authors of this review identified 71 case-control studies, cohort studies, and randomized controlled trials (RCTs) assessing the effectiveness of influenza vaccination against influenza or influenza-like illness; 64 of these studies addressed vaccine effectiveness in older persons.

There were five RCTs addressing vaccine effectiveness, with about 5,000 observations. Because vaccine type, setting, and outcomes were different in each study, conclusions were limited. A meta-analysis of two of these trials with a total of 2,567 patients showed that inactivated vaccines were more effective than placebo against influenza-like illness and influenza in community settings where there is a high viral circulation (vaccine effectiveness [VE] = 43 percent; 95% confidence interval [CI], 21 to 58 percent); pooled data from three RCTs showed that the vaccines were effective against influenza (VE = 58 percent; 95% CI, 34 to 73 percent).

Vaccines appear to be safe. Based on pooled data from four RCTs, local adverse events such as tenderness and sore arm were significantly more common with vaccine than with placebo; however, differences in systemic side effects were not statistically significant between the two groups.

In the cohort studies, healthier patients with better access to health care services may have received the vaccination preferentially, making the comparison groups different at baseline and overestimating the effectiveness of the vaccine. Likewise, the benefits for all-cause mortality found only in observational studies may reflect differences between vaccinated and nonvaccinated groups rather than a true effect. Evidence from RCTs, in which bias is reduced, is limited because of the relatively small total number of patients studied. Because of universal recommendations for influenza vaccination in older persons, it is unlikely that a randomized placebo-controlled trial will be performed to provide the high-quality evidence required.

Overall, these results support the conclusion that vaccination benefit is maximized if those at greatest risk for influenza complications are successfully targeted for vaccination. The effectiveness of vaccination in older persons living in community-based settings was more modest. The Advisory Committee on Immunization Practices recommends annual influenza vaccination for the following groups: persons at high risk for influenza-related complications and severe disease, including children six to 59 months of age, pregnant women, persons 50 years or older, and persons of any age with certain chronic medical conditions; and those who live with or care for persons at high risk.1