Journal of the American College of Nutrition, Vol. 26, No. 90005, 570S-574S (2007)
Published by the American College of Nutrition
How Primary Care Providers Might Review Evidence on Hydration
Kathryn M. Kolasa, PhD, RD, LDN,
Carolyn J. Lackey, PhD and
David G. Weismiller, MD, MSc
Brody School of Medicine at East Carolina University (K.M.K.), Brody School of Medicine at East Carolina University (D.G.W.), Greenville, North Carolina State University, Cooperative Extension Service (C.J.L.), Raleigh, North Carolina
Address correspondence to: Kathryn Kolasa PhD, 4N70 Brody, East Carolina University, Greenville NC 27834. E-mail: kolasaka{at}ecu.edu
 |
ABSTRACT
|
|---|
Primary care providers (PCPs) are increasing their use of evidence-based medicine (EBM) in the care they give patients. They evaluate the available evidence to determine if it applies to their patients and seek to complement their clinical experience with EBM to improve patient outcomes. In evidence-based practices, patient oriented data are valued more highly than disease oriented evidence. More than 8 million biomedical articles are published annually, but only an estimated 2% of those are relevant to improved patient outcomes (POEMs - patient oriented evidence that matters). This paper describes some of the tools used by PCPs to search for evidence and the decision-making process used to determine if they will change their practice. Understanding how PCPs evaluate research findings and other evidence can help hydration researchers frame their research questions and study reports.
Key words: evidence-based medicine (EBM), information mastery, patient oriented evidence that matters (POEMS)
Key teaching points:
• 2% of published biomedical articles are relevant to improved patient outcomes.
• Primary care providers (PCPs) use standard techniques for evaluating the usefulness of research reports. Understanding these tools may help hydration researchers frame their research questions and study reports.
• A variety of systems are used to grade evidence.
• Increasingly electronic data bases are queried at the point of care.
|
|
INTRODUCTION
|
|---|
This paper describes ways primary care providers (PCPs) use electronic data sources to evaluate research information to make evidence-based patient-oriented clinical recommendations. The term "evidence-based" is defined differently in various disciplines. Therefore, researchers who want to impact recommendations PCPs make about fluid intake, need to understand the terms and processes clinicians use to evaluate available evidence.
The emphasis on evidence-based practice is partially driven by health care providers striving to meet the goal of helping individuals of genders, all ages and ethnicities to increase life expectancy and improve quality of life [1]. Also, PCPs are challenged to become skilled in evaluating the wealth of available clinical data and using electronically available tools in the evaluation process. This is often referred to as information mastery [2].
|
|
PREPARING FOR THE SEARCH
|
|---|
As part of information mastery, clinicians faced with "knowledge needs" from either their inpatient or outpatient encounters translate their questions into an answerable clinical question using the PICO format. In PICO the P stands for patient or problem; I for intervention; C for comparison intervention; and O for outcomes [3]. Once the question is framed, the clinician seeks answers by asking a colleague, checking a text book, or searching the literature to find the best available evidence. It is important to recognize that the best available evidence may take on many forms: published research report, clinically relevant summary, or a consensus paper. Regardless of the type of report, the clinician is optimally looking for what is referred to as Patient-Oriented Evidence that Matters (POEMs). POEMs address questions that clinicians face; measure outcomes PCPs and patients care about (e.g. symptoms, morbidity, quality of life, and mortality) and have the potential to change practice. This is in contrast to a report of disease oriented evidence (DOE). Most basic science research reports are DOEs. Fig. 1 shows the hierarchy of evidence.
Although DOE is important, hydration researchers need to understand that the PCP has a much greater interest in the POEM evidence and may not read reports of DOE. Additionally, the PCP considers the strength of the evidence. There are more than 100 evidence grading systems in use. One example is that used by The American Academy of Family Physicians. Reviewers using that system assign an "A" when the recommendation is based on consistent, good quality patient oriented evidence; a "B", when the evidence is inconsistent or limited quality patient oriented; and a "C" for consensus, disease oriented evidence, usual practice, expert opinion, or case series for studies of diagnosis, treatment, prevention, or screening [4]. Another example is that used by the US Preventive Health Service, which rates clinical recommendations as: strongly recommends, recommends, not for or against, against, and insufficient. The American Dietetic Association maintains an online evidence analysis library (www.adaevidencelibrary.com) of nutrition research. It uses grades I–V to indicate the overall strength or weakness of evidence informing conclusion statements.
|
|
DECIDING ON THE ELECTRONIC TOOL TO USE
|
|---|
The increasing availability of computers in the exam room or at the patient's bedside allows the PCP to access information quickly. Traditionally, PCPs with a question might have searched Medline for a research paper outside of the patient care time, studied the paper, and decided if the study was valid and applicable to their practice. Today's PCPs have learned in medical school, residency and in continuing education how to drill for the best information (Fig. 2). PCPs skilled in information mastery are likely to search electronic secondary sources first, looking for relevant information evidence that may change their practice at the point of care. We will describe some of the websites that clinicians frequently browse: Cochrane Collaboration (www.cochrane.org), InfoPOEMS (www.infopoems.com), MD Consult (www.mdconsult.com), Medline (www.pubmed.gov), Google (www.google.com) and HighWire Press (www.highwire.stanford.edu).
The Cochrane Database of Systematic Reviews electronic library is updated quarterly and is produced by the Cochrane Collaboration. It produces exhaustive systematic reviews of health care intervention trials and also offers databases of controlled clinical trials addressing prevention, treatment, diagnosis, and screening. The authors use both published and unpublished findings and follow a rigorous and explicit process including peer review. PCPs believe that Cochrane is complete, reliable, and practical and the "Gold Standard" for information mastery. Cochrane is at the tip of the evidence pyramid because it has the power to detect effectiveness or harm that may be missed in smaller studies, it invites comments and criticisms, and it outlines implications for clinical practice or research. The weaknesses include potential for selection bias, attempts to combine data from disparate studies, and "negative" studies may not be available. The absence of a systematic review matching a PCPs needs does not mean that evidence to support a practice is not available, only that a review on that topic has not been conducted. Instructions for potential authors of systematic reviews are available at www.cochrane.org. The Cochrane Primary Health Care Field is coordinated by Dr. Lorne Becker, Department of Family Medicine at SUNY Upstate Medical University.
Specialty specific POEMS are near the top of the evidence pyramid. They are short abstracts with reviews and are written in a standard format including the clinical question, bottom line, study design, funding, setting, and synopsis. Preparing POEMS may be an efficient strategy for hydration researchers to counter widely held myths. For example, the group Inforetriever sends daily e-mails to subscribers in the POEMS family medicine specialty. An archive is also easily accessible (www.infopoems.com).
There are several information aggregators such as MD Consult, PubMed Central and Ovid. These are collections of articles published by various publishers and reformatted to allow standardized searching and other functionality. On a typical day, a PCP in our practice may seek the answer to a clinical question first through UpToDate and then MD Consult. Each is easily accessible through our electronic medical record. The Usefulness of a particular data base is judged by the relevance times the validity divided by the amount of work to get it (Fig. 2).
Medline is, of course, a familiar tool to hydration researchers. More than 4,800 biomedical journals with greater than 12 million citations are available online.
Search engines index large sections of the internet. The cached content can then be searched using sophisticated algorithms. Google Scholar searches the scholarly literature. Yahoo provides direct access to bibliographic information, abstracts, and online articles.
Portal sites create an interconnected content network that can often be searched through a single interface. HighWire Press, ScienceDirect, and Wiley InterScience are examples. The High Wire portal (www.highwire.stanford.edu), established at Stanford University, has a mission to help society publishers compete in the online world. It hosts high impact, peer-reviewed, scholarly content including a large archive of full text articles and continuing medical education (CME) sites. Hydration researchers may consider publishing their work in these journals.
Those PCPs practicing evidence-based medicine seek to change their practice behavior based on POEMs. They believe this will improve patient treatments and outcomes. A helpful web-based course on EBM is sponsored by Michigan State University (http://www.poems.msu.edu/InfoMastery/). Since 2002, the American Academy of Family Physicians has been providing incentives to both presenters and participants of CME courses to be evidence-based. Table 1 lists selected EBM Resources.
|
|
AN EXAMPLE
|
|---|
To illustrate how a PCP in our program might seek evidence-based knowledge for his or her practice, we will explore the case of a 65 year old woman with a history of recurring kidney stones who asks her physician, "How much water do I need to drink every day to prevent the recurrence of stones?" Using the PICO format, the patient's question would look like this: P (population)-in women over the age of 65 years with a history of kidney stones; I (intervention)-consuming more water/fluid daily compared to average consumption; and CO (clinical outcomes)-reduce the incidence of kidney stones? The physician might ask his dietitian team member for advice and the coworker replies that the Tufts Food Guide Pyramid for Older Adults recommends eight or more servings of water/liquid. With a little time on his hands, the physician might decide to do a quick search of reviews. Using the terms geriatrics, kidney stones, water consumption, and fluid consumption, a search of the Family Physicians Inquiries Network (FPIN) archives and INFOPOEMS identified no recommendations. However, there was a "five minute clinical consult overview." At the level of "expert opinion" evidence, one strategy outlined was a fluid intake of 2 L/day, or 8 eight-oz glasses/day besides meals to reduce the likelihood of recurrence of kidney stones.
Some PCPs would stop there and provide that advice to the patient. However, if the PCP wanted to know if there was any stronger evidence than "expert opinion," he may decide to do a bit more work and continue searching electronic resources. In this case, he searched: Cochrane, MD Consult, Medline via Ovid, Google and HighWire.
A search of MD Consult identified a paper entitled "Medical treatment of nephrolithiasis" [5]. This review article discussed increased urinary volume and identified no long-term studies of the effects of different beverages. It did provide a schematic practical guideline for medical treatments (water therapy plus diet). The Medline search identified two papers. The first was a review paper entitled "Body weight, diet and water intake in preventing stone disease" [6]. The authors did not grade the evidence but suggested that a recommendation of more than two liters of water per day would reduce risks by 50% over five years. This review had three relevant citations that were posted from the 1990s. The second paper was a review of kidney stones by Parmar [7] based on a Medline search for the period 1990-2003. Two points relevant to the case were made: (1) stones are associated with hypertension and low magnesium intake in post menopausal women, and (2) fluid intake should be increased to maintain urine output of two to three liters/day. It noted that fluid alone would not reduce stones in patients with hypercalciuria.
Cochrane had a review entitled "Water for preventing urinary calculi" updated November 2005 [8]. The "plain language summary" is that increased water intake may help reduce the risk of recurrence of kidney stones but more studies are needed. It noted that future trials should be designed to evaluate how much water and what kind of water is best for the primary and secondary prevention of urinary calculi.
Surprisingly, the search of Medline via Ovid identified no papers. Google led to the American Kidney Fund website [9] where the following advice was given:
"Try to drink 8 to 12 full glasses of water every day. Drinking lots of water helps to flush away the things that form stones in the kidney. You can also drink ginger ale, lemon-lime sodas, and fruit juices, although water is best. Limit your coffee, tea, and cola to one or two cups a day because the caffeine may cause you to lose fluid too quickly".
A search of the HighWire Press database yielded two papers. The first was a 1996 report of a prospective study of the relation between the intake of 21 different beverages and the risk of symptomatic kidney stones in a cohort of 45,289 men ages 40–75 years, who had no history of kidney stones [10]. The second paper, a 1998 prospective study of 81,093 women ages 40–65 years, seemed more relevant [11]. The researchers concluded that "an increase in total fluid intake can reduce risk for kidney stones, and the choice of beverages may be meaningful."
The search of all these sites took about 20 minutes by the PCP.
|
|
DETERMINING THE BEST EVIDENCE FOR THE CASE AT HAND
|
|---|
As they design their studies or prepare their reports, hydration researchers might benefit from understanding how PCPs determine the best evidence and then how they evaluate the research reports. PCPs are taught in medical school or residency to use worksheets [12] to guide them through the critical appraisal of a paper or guideline. With practice, the questions become second nature. For example, if the best evidence found is a treatment article, the PCP needs to determine the relevance to the patient and will first ask, "Is the study outcome something the patient cares about?" If the answer is "no," the PCP won’t read the paper and will move on. If the answer is "yes," a PCP will determine the study's validity by asking a series of questions such as: - "Was the study a controlled trial with patients randomly assigned?"
- "Are the patients so dissimilar to mine that the results don’t apply?"
- "Were all patients accounted for at the conclusion of the trial?"
- "Were patients and study personnel blinded?"
- "Were intervention and study groups similar?"
- "Are the results clinically significant?"
- "If it was a negative trial, was the power of the study adequate?"
If it was a review article, the questions might include: - "Was most of the relevant research included?"
- "Did the authors attempt to critically evaluate the research?"
- "Was the quantitative information presented appropriate and complete?"
- "Was the discussion balanced and free from bias?"
- "Were the references appropriate and current?"
- "Were the conclusions valid and opinions noted as such?"
If the best evidence was a clinical practice guideline, the PCP might ask these questions: - "What was the quality of the evidence?‘
- "Was the guideline based on at least one properly designed randomly controlled trial (RCT)?‘
- "If not, was it based on a well-designed nonrandomized controlled trial or a well-designed cohort or case control study, preferably conducted at more than one center; or based on dramatic results in uncontrolled experiments?‘
- "If not, was it based on the opinions of respected authorities?‘
The validity of the guideline would be assessed by determining if the report had a balance sheet, evidence tables, or other indications of supporting evidence strength. The PCP would want to determine if an explicit and sensible process was used to identify, select, and combine evidence; and if the guideline authors considered the relative value of the different outcomes. Additionally the PCP would want to know if the guideline had been subjected to peer review and/or testing.
In addition to the questions posed above, PCPs like to see reports that have strong abstracts and when possible give relative and absolute risk reduction, numbers needed to treat, numbers needed to harm, likelihood ratios, and predictive values.
Even with all the tools, PCPs may not find the exact answers to the clinical questions but the unanswered questions can be the basis for further research. In the meantime, PCPs take the evidence and blend it with their clinical experience to make recommendations to their patients [13].
|
|
FOOTNOTES
|
|---|
Conflict of Interest Disclosure: This review was funded by the North American branch of the International Life Sciences Institute. The author declares that no present corporate or government relationship involves a conflict of interest, with regard to this manuscript. Drs. Lackey and Weismiller have no conflicts of interest to declare in connection with this work.
Received July 16, 2007.
|
|
REFERENCES
|
|---|
- US Department of Health and Human Services:"Healthy People 2010,"
2nd ed. Washington, DC:2000
. Available from: www.healthypeople.gov
- Ebell M:An Introduction to Information Mastery
. Available from: www.poems.msu.edu/InfoMastery.
- Ely JW, Osheroff JA, Ebell MH, Bergus GR, Levy BT, Chambliss ML, Evans ER: Analysis of questions asked by family doctors regarding patient care.BMJ319
:358
–361,1999
.[Abstract/Free Full Text]
- American Academy of Family Physicians: Strength of recommendation taxonomy (SORT).Am Fam Phys74
:1666
–1667,2006
.
- Borghi L, Meschi T, Schianchi T, Allegri F, Guerra A, Maggiore U, Novarini A: Medical treatment of nephrolithiasis.Endocrinol Metab Clin North Am31
:1051
–1064,2004
.
- Meschi T, Schianchi T, Ridolo E, Adorni G, Allegri F, Guerra A, Novarini A, Borghi L: Body weight, diet and water intake in preventing stone disease.Urol Int72 (Suppl 1)
:29
–33,2004
.
- Parmar MS: Kidney stones.BMJ328
:1420
–1424,2004
.[Free Full Text]
- Qiang W, Ke Z: Water for preventing urinary calculi. The Cochrane Database of Systematic Reviews. Accession Number: 00075320-100000000-03314, November 23,2005
.
- American Kidney Fund: Kidney Stones. Rockville, MD: American Kidney Fund, 2007. Available from: www.kidneyfund.org
- Curhan GC, Willett WC, Rimm EB, Spiegelman D, Stampfer MJ: Prospective study of beverage use and the risk of kidney stones.Am J Epidemiol143
:240
–247,1996
.[Abstract/Free Full Text]
- Curhan GC, Willett WC, Speizer FE, Stampfer MJ: Beverage use and risk for kidney stones in women.Ann Intern Med128
:534
–540,1998
.[Abstract/Free Full Text]
- Ebell M: Information at the point of care: answering clinical questions.J Am Board Fam Pract12
:225
–235,1999
.[Medline]
- Shaughnessy AF, Slawson DC, Becker L: Clinical jazz: harmonizing clinical experience and evidence-based medicine.Fam Practice47
:425
–428,1998
.
TOP
ABSTRACT
INTRODUCTION
