Some evidence-based cases

Case study 1: persistent cough

In this section we will discuss several case studies that show how EBP can help in a range of clinical situations. You can then think of a clinical question of your own and we will try to answer it.

A 58 year old who was visiting her !P a"out another #atter said$ as an aside$ %&an you do anything a"out a cough'( )he had had a persistent cough for *+ years with various treat#ents "ut no cure. )he had "een referred twice to physicians. ,h e !P searched Pu"-ed .the we" "ased version of -E/0I1E2 using %&linical 3ueries($ which is a category of Pu"-ed designed for clinicians .see pages 545582. ,h e search for persistent cough revealed that the #ost co##on causes are6 7 postnasal drip 7 asth#a 7 chronic "ronchitis. ,h e !P thought the cough was #ost likely to "e due to asth#a$ and prescri"ed appropriate fi rst line treat#ent. ,h e patient thought she had already tried that treat#ent and that it did not work "ut tried it again anyway$ without success. 8owever$ the search also showed that gastro oesophageal refl u9 is a less co##on "ut possi"le cause of persistent cough .:+; of cases2$ which the !P had not known "efore. ,h e !P therefore reco##ended the patient to take antacids at night and raise the head of her "ed. After one week$ her cough disappeared for the fi rst ti#e in *+ years and has not co#e "ack since.

How did EBP help?

,h is case raises interesting questions of what doctors %should( know. It was written up in the B-< and pu"lished as an e9a#ple of how EBP can help !Ps. 8owever$ so#e physicians wrote in saying that %everyone should know( that gastro oesophageal refl u9 was a possi"le cause of cough. ,h e author replied that although respiratory physicians #ight know this infor#ation$ !Ps did not necessarily know it. An anaesthetist wrote in to say that after reading the article he had "een treated for gastro oesophageal refl u9$ which had cured a cough he had had for =+ years> Conclusion: EBP can help you find the infor#ation you need$ whether or not you %should( already know it.

Reference:
!las?iou P .:@@82. Evidence "ased case report6 ,wenty year cough in a non s#oker. British Medical Journal =:46:44+5:44:

Case study 2: dog bite

A patient ca#e to the clinic with a fresh dog "ite. It looked clean and the !P and patient wondered whether it was necessary to give prophylactic anti"iotics. ,h e !P searched -E/0I1E and found a #eta analysis indicating that the average infection rate for dog "ites was :A; and that anti"iotics halved this risk. In other words6 7 for every :++ people with dog "ites$ treat#ent with anti"iotics will save B fro# "eco#ing infectedC or 7 treating :A people with dog "ites will prevent one infection. ,h e second nu#"er .:A2 is called the %nu#"er needed to treat( .11,2. ,h e !P e9plained these fi gures to the patient$ along with the possi"le consequences of an infection$ and the patient decided not to take anti"iotics. Dn follow up$ it was found that he did not get infected.

How did EBP help?

In this case$ EBP helped "ecause the e#pirical data were easy for the patient to understand and he could participate in the clinical decision. As the culture of health care changes further towards consu#er participation in health care decision #aking$ patients will de#and this type of infor#ation.

Reference:

&u##ings P .:@@A2. Anti"iotics to prevent infection in patients with dog "ite wounds6 a #etaanalysis of rando#i?ed trials. Annals of Emergency Medicine *=65=555A+.

Case study 3: microscopic blood in the urine

Dne of us$ then a healthy AB year old #ale$ was acting as a patient in a #edical e9a#. ,h e students accurately found #icroscopic traces of "lood in his urine. 8e went to his !P and was retested a #onth later. ,h e "lood was still there. ,h e !P suggested conventional investigation6 an ultrasound and cystoscopy. It was ti#e to search the literature for evidence of the eff ectiveness of these procedures. 8e searched for a cohort study of A+55+ year olds with hae#aturia with longter# follow up and for E&,s of screening for hae#aturia. 8e used the search categories %prognosis( and %specifi city( and the search ter#s %hae#aturia DE he#aturia(. 8e got =++ hits. ,wo papers were very relevant .see "o92. ,h erefore$ he concluded that "lood in urine is not a good indicator of "ladder cancer and did not have the cystoscopy test.

How did EBP help?

,h e lesson fro# this case concerns the practical versus the e#pirical. /octors tend to think along the lines of6
Blood does not "elong in the urine so it #ust "e co#ing fro# so#ewhere. It could "e co#ing fro# a potentially serious cause$ such as "ladder cancer.

E#pirical questions$ on the other hand$ ask a"out outco#es F in this case$ whether conventional investigation leads to "etter health outco#es. 8ere$ the evidence .surprisingly2 showed that such investigation provides no "enefi t$ "ecause #icroscopic hae#aturia see#s to "e no #ore prevalent a#ong those who later develop urological cancer than those who do not. Dnce again$ "eing e#pirical and quantitative allows patients to participate #uch #ore fully in clinical decisions.

Study 1
:+$+++ #en were screened. A"out *5+ .*.5;2 had hae#aturia. ,h ese #en were asked to visit their !P and a"out :5+ .4+;2 did so. Df those$ only three had a serious pro"le#. Df these6 7 * had "ladder cancer 7 : had refl u9 nephropathy. ,h is shows that there is a"out a : in 5+ chance of having a serious disease.

Study 2
As part of a personal health appraisal$ *+$+++ #en were given a urine test. Gollow up studies of the #en who were positive for hae#aturia found three cancers per year$ or :.5 cancers per :+++ person years. 8owever$ the people who did not have hae#aturia were also followed up and the rate of cancer for these people was

e9actly the sa#e as for the people with hae#aturia.

Reference:
/el -ar & .*+++2. Asy#pto#atic hae#aturia H in the doctor. British Medical Journal =*+6:455:44.

EBP can help to reduce litigation

,h is case raises the issue of possi"le litigation. Ihat if the patient is not tested and later develops a serious disease' 8owever$ "ecause EBP i#proves co##unication "etween doctors and patients and allows patients to share decision #aking$ it protects doctors fro# litigation ."ecause #ost litigation happens when there is a "reakdown in co##unication2. EBP analyses have already "een used in the courts and have "een well accepted. )uch e#pirical evidence has saved doctors fro# trou"le when opinion #ay have da#ned the#.