The Hong Kong Practitioner

Quality Assurance and Accreditation Committee	*The Hong Kong Practitioner* VOLUME 26 / February 2004

S a m p l e

Continuous professional development (CPD)

Evidence Based Practice and CPD accreditation

Search on medical literature to answer one's clinical questions is an efficient self-directed and patient-oriented option in continuous medical education. Since 2002, up to 5 CPD points have been accredited on each Critical Appraisal Report. A new stem on the generation of a scenario-based clinical question and some modifications are made in the new worksheets to make the process of learning more clinically focused. To encompass this approach, the name of the exercise is changed from "Critical Appraisal Report" to "Evidence Based Practice"

On top of the previous Therapeutic Intervention checklist, another one on Diagnostic Test is now enclosed. Two examples below are employed to illustrate the use of current best evidence to improve our patient care. The original articles are freely accessible at

http://bmj.bmjjournals.com/cgi/content/full/321/7266/931
http://bmj.bmjjournals.com/cgi/content/full/327/7424/1144

Evidence Based Practice on Therapeutic Intervention

Name : Dr. QAA

Description of the Clinical Scenario that generates your clinical question:

Mrs. Wong brought her 4-month old infant for his second DPT vaccination. She told me that following the first vaccination; the skin around the injection site had been very red and swollen for several days. I am not sure if the size of the needle would affect local reactions.

Your Clinical Question:

When giving immunization injections to infants, does needle size relate to the local complications at the injection site?

Title of literature that best answer your clinical question:

Effect of needle length on incidence of local reactions to routine immunization in infants aged 4 months: randomized controlled trial.

Published in:

Hong Kong Practitioner	:	Vol	No	Page

Or	:	BMJ 2000; 321:931-933

		N.B. An authorized copy of the article should be enclosed if the article is taken from outside the Hong Kong Practitioner.

Critical Appraisal Checklist:

What is the intervention?

Immunization with 23 gauge, 25 mm, blue hub needle vs 25 gauge, 16 mm, orange hub needle

Were the results valid?

(a)	Were groups of patients randomized?
	Yes	No	Don't know
(b)	Was randomization list concealed?
	Yes	No	Don't know
(c)	Were the intervention and control groups similar at the start of the trial?
	Yes	No	Don't know
(d)	Were patients & investigators blinded?
	Yes	No	Don't know
(e)	Was follow-up complete?
	Yes	No	Don't know
(f)	Were patients analyzed in the groups to which they were initially randomized?
	Yes	No	Don't know
(g)	Were there clear measures of outcome?
	Yes	No	Don't know

What is your comment on the validity of the result?

Of the 119 infants recruited, randomization was well performed by concealed allocations via opaque envelopes. Baseline characteristics of both groups were very similar. Follow up was complete with nurse ringing up the parent if there was a delay in the return of report of reactions. The loss to follow up was clearly stated. However, there was no blinding in the assessors, i.e. the parents, who witnessed the injection process. Overall, this study is of good validity.

What are the results?

(a)

How large was the treatment effect?

Outcome	Absolute Risk Reduction (ARR)	Relative Risk Reduction (RRR)	Number Needed to Treat (NNT)
Redness at 1 day	0.35	0.55	2.9
Redness at 3 days	0.24	0.85	4.2
Swelling at 3 days	0.27	0.68	3.7
Any local reaction	0.22	0.26	4.5

(b)

How precise was the estimate of the treatment effect?

Judging from the confidence interval quoted in the article's result table, the results were acceptable in the precision.

What is your comment on the result?

On average, for every 5 infants vaccinated, use of the 23 gauge, 25 mm needle would prevent 1 infant from redness or swelling at day 3.

Are the results applicable to your patients?

(a)	Are the subjects in the study similar to yours?
	Yes	No	Don't know
(b)	Does the result have a potential benefit/harm on the health of your patients?
	Yes	No	Don't know
(c)	Is the intervention feasible and available to your practice?
	Yes	No	Don't know

What is your comment on the applicability of the result?

It is feasible in my clinic to employ the longer needle, i.e. 23 gauge, 25mm when giving routine immunization injections to infants. However, one possible harm is deep penetration in lean infants, with the nightmare of periosteal or bone injection!

How are you going to apply the evidence to your practice?

I shall use the longer needles to give routine vaccination to infants at or above the age of 4 months, ensuring an intramuscular injection. However, for the 0 to 4 months or lean babies, I would be more confident to use the shorter ones.

Other learning points you may wish to discuss/comment:

The method of injection used was anterolateral thigh, stretching the skin taut and inserting the needle at a 90° angle to the skin, with the needle inserted into the skin up to the hub.

The needles used in the study differed in both length (16 v 25 mm) and bore (25 v 23 gauge), we cannot know which of these factors determined the observed differences in the rates of redness and swelling.

For Official Use Only

Score

Name of Assessor

Date of Assessment

Comment from Assessor

S a m p l e

Evidence Based Practice on Diagnostic Test

Name : Dr. QAB

Description of the Clinical Scenario that generates your clinical question:

Day in and day out, I often see patients with myriads of psychosocial complaints. I am not sure if simply asking about depressed mood and loss of interest good enough in picking up depressed patients in my practice.

Your Clinical Question:

How accurate are asking about depressed mood and loss of interest in diagnosing depression in a primary care setting?

Title of literature that best answer your clinical question:

Screening for depression in primary care with two verbally asked questions: cross sectional study.

Published in:

Hong Kong Practitioner	:	Vol	No	Page

Or	:	BMJ 2000; 321:931-933

		N.B. An authorized copy of the article should be enclosed if the article is taken from outside the Hong Kong Practitioner.

Critical Appraisal Checklist:

What is the diagnostic test?

By asking 2 questions:
a.	During the past month have you often been bothered by feeling down, depressed, or hopeless?
b.	During the past month have you often been bothered by little interest or pleasure in doing things?

Were the results valid?

(a)	Was the diagnostic test evaluated in an appropriate spectrum of patients?
	Yes	No	Don't know
(b)	Was there an independent, blind comparison with a reference ("gold") standard of diagnosis?
	Yes	No	Don't know
(c)	Was the reference standard applied regardless of the index test result?
	Yes	No	Don't know
(d)	Was the test validated in a second, independent group of patients?
	Yes	No	Don't know
(e)	Were the methods for performing the test described in sufficient detail to permit replication?
	Yes	No	Don't know

What is your comment on the validity of the result?

In this study, there is no non-screened group to act as the "control". Both the "test" and the "gold standard" were not clearly stated. e.g. the 2 questions that were asked actually had 5 components. How the doctor delivered the message verbally may not be standardized.

What are the results?

	With the disease - Depressed	Without the disease - Not depressed
Test positive - "Yes" to either questions	28	129	157
Test negative - "No" to both questions	1	263	264
	29	392	421

a.	Sensitivity	[28/29]	96.6%
b.	Specificity	[263/392]	67.1%
c.	Positive predictive value	[28/157]	17.8%
d.	Negative predictive value	[263/264]	99.6%
e.	Likelihood ratio for a positive test result	[96.6/(100-67.1)]	2.93
f.	Likelihood ratio for a negative test result	[(100-96.6)/67.1]	0.05

What is your comment on the result?

The 2 verbally asked questions have good sensitivity but fair specificity. If a patient gives a "yes" to both questions, the chance that it is a false positive is nearly 5 times of a true positive (129 vs 28; PPV of 17.8%).

The results suggest that the questions are more useful in excluding depression. Only 1 false negative was found in this study (1 vs 263; NPV of 99.6%). In one local study, the prevalence of suicidal ideation amongst elderies in Hong Kong is 6%. Applying the likelihood ratio normogram, with a likelihood ratio for a negative test of 0.05, a patient saying "no" to both questions would have a 0.2% chance of being depressed.

Are the results applicable to your patients?

(a)	Is the test available, affordable, accurate and precise in your clinic setting?
	Yes	No	Don't know
(b)	Will the results change your management strategy?
	Yes	No	Don't know
(c)	Will patients be better off as a result of the test?
	Yes	No	Don't know

What is your comment on the applicability of the result?

The questions are affordable and accurate as a negative screening tool. However the questions need further validation in Cantonese yet.

How are you going to apply the evidence to your practice?

I would have this piece of evidence in mind when I next suspect a patient having depression. I would be more confident in ruling out depression if the patient gives "no" to both questions. But if the answer is "yes" to either one, I know I might have to do a more thorough workup with the patient.

Other learning points you may wish to discuss/comment:

Given more evidence, it is attractive to use a short and quick aid in the hypothesis testing our daily consultations. As Chris Del mar comments in the editorial, we are at the dawn of a new phase of evidence-based practice. By having more research on similar areas, hopefully we can separate the wheat from the chaff, and have our clinical work further simplified.

For Official Use Only

Score

Name of Assessor

Date of Assessment

Comment from Assessor

i.	Yip PS et al. A prevalent study of suicide ideation among older adults in HK SAR. Int J Geriatr Psychiatry 2003 Nov;18(11):1056-1062.
ii.	D Sachett et al. How to practice and teach EBM 2ed. P.79.