10 CHAPTER1 Geriatric Physical Therapy in the 21st Century B0X1-2 Relevant a clinica Joumal of the American Geriatric Society ent or situ to your Medical and Biological Sciences Credibility.Searching for credible evidence starts ouus Physical Therapy inthe procyprovide the highest level ofv dence. alidit of the table of contents and carefully select a small number potential studies.Regardless of its general category (i.e. tematic review),the dence,th study wer ngs and reporte of th A second approach is to go to a site such as AMEDEO com).This is a free service providing textbooks A brief summary of selected points is aggrega alons specific provided below care special- studies compare the ld man specialty area (or a subset of these journals as requested) should confirm represent iveness of the subjects in the free PubMed account tha ows a str run periodically to identify any new citations,and have findings of the other to avoid any biasing influence,all the new citations automatically forwarded via e-mail. subjects should undergo the gold standard,and,ideally. cif bMed approach all the the study shoul repeated with a new set of subjects 0 onhrm the with Sources that Translate Evidence into Practice of the outcome of interest.Prognosis studies may follow either one or two groups atients (cohort or case and erature im sources for translatine evidence into of prognosis studies inform particularly on a systemat rev v of the literatu length of tim expert cons re validity dpaicemateicieto the gth of follow-uD. and pro spective desi HealthCare Rese assembled at a common point in the Department of Hea rvice provides a of the di 5 are subjects reason theng me data en e cht at this poin easc,an out larg which the guideline is based.Strength of the evidence nced the predicted outcome?Wer re the outcomes criteria should be based on quality,consist ncy,and number of ree studies supporting the recommendation biases and responsive Critically Assessing the Evidence characteristics or condition of interest in the study.Kev "Best available evidence nce of a a somparison roup to whish ubest of a control ilarity at baseline and low attritiot (collected,analyzed,and reported using unbiased and the study.The methods used in the study should mini- valid processes),is clinically important (the study's mize risk of researcher bias or confounding variables10 CHAPTER 1 Geriatric Physical Therapy in the 21st Century table of contents and carefully select a small number of particularly applicable articles to read full-text. The higher the article is on the pyramid of evidence, the more likely its findings can be readily applied to clinical practice. A second approach is to go to a site such as AMEDEO (www.amedeo.com). This is a free service providing weekly e-mails aggregating article citations specific to any interest across a wide range of health care special￾ties. The citations are typically taken from ongoing searches of newly published issues of core journals in the specialty area (or a subset of these journals as requested). A third option is to set up a free PubMed account that allows an individual to identify and save a specific search strategy within PubMed, have the search automatically run periodically to identify any new citations, and have the new citations automatically forwarded via e-mail. The PubMed approach allows you to be the most spe￾cific about the characteristics of the studies of interest and searches across the widest variety of journals. Sources that Translate Evidence into Practice Recommendations. Systematic reviews that provide evidence of objective and unbiased synthesis of the full body of literature on a topic, providing unambiguous and well-grounded recommendations, are important sources for translating evidence into practice recom￾mendations. Clinical practice guidelines, particularly those based on a systematic review of the literature and expert consensus in applying the evidence to clini￾cal practice, can be efficient sources of evidence. The National Guideline Clearinghouse of the Agency for HealthCare Research and Quality (AHRQ) of the U.S. Department of Health and Human Services provides a central and searchable guideline database. When exam￾ining the Practice Guidelines, confirm the comprehen￾siveness and objective analysis of the literature on which the guideline is based. Strength of the evidence should be based on quality, consistency, and number of studies supporting the recommendation. Critically Assessing the Evidence “Best available evidence” has become a catch phrase to describe preferred information sources for evidence￾informed practice. But what exactly does best evidence really mean? Best evidence is evidence that is credible (collected, analyzed, and reported using unbiased and valid processes), is clinically important (the study’s findings demonstrate a large enough change to have a clinically meaningful impact), and the credible and important findings are directly applicable to your patient or situation. Credibility. Searching for credible evidence starts out us￾ing the procedures described in the previous section to locate studies likely to provide the highest level of evi￾dence. Credibility (quality) is assessed through a critical assessment of the internal and external validity of the potential studies. Regardless of its general category (i.e., therapy, prognosis, diagnosis, or systematic review), the study should provide convincing evidence that data were collected, analyzed, and reported in an unbiased fashion. A full review of the concepts of critical assessment of the biomedical literature is available in several well-organized textbooks.2,15,16 A brief summary of selected points is provided below. Diagnosis studies compare the performance of a new diagnostic test against the current gold standard or its equivalent,17 typically testing the test. Diagnosis studies should confirm representativeness of the subjects in the study and present a solid argument that justifies the choice of gold standard.18 Assessors for reference and target tests should be independent and blinded to the findings of the other to avoid any biasing influence, all subjects should undergo the gold standard, and, ideally, the study should be repeated with a new set of subjects to confirm the findings. Prognosis studies follow subjects with a target disor￾der or risk factor over time and monitor the occurrence of the outcome of interest. Prognosis studies may follow either one or two groups of patients (cohort or case– control, respectively), preferably prospectively, to exam￾ine the impact of various factors on the target disorder. The findings of prognosis studies inform judgments about such things as who is most likely to benefit from rehabilitation or the length of time to achieve rehabilita￾tion goals. Key indicators of credibility and validity of prognosis studies19 include the representativeness of the subjects, length of follow-up, and prospective design. Were subjects assembled at a common point in the course of the disease, are subjects reasonably representa￾tive of the typical patient at this point in the disease, and are subjects followed for a sufficiently long time period, without large attrition, to capture everyone who experi￾enced the predicted outcome? Were the outcomes criteria free of patient or practitioner biases and responsive enough to capture the outcome if it occurred? Therapy studies assess the impact of specific interven￾tions on subjects chosen because they possess the specific characteristics or condition of interest in the study. Key indicators of quality in a therapy study are the presence of a control or a comparison group to which subjects were randomly assigned, reasonable between-group sim￾ilarity at baseline, and low attrition over the course of the study. The methods used in the study should mini￾mize risk of researcher bias or confounding variables BOX 1-2 Key Journals Particularly Relevant to Geriatric Physical Therapy Journal of the American Geriatric Society Journal of Gerontology: Series A; Medical and Biological Sciences Journal of Geriatric Physical Therapy Physical Therapy