-
Notifications
You must be signed in to change notification settings - Fork 2
Expand file tree
/
Copy pathmethods.html
More file actions
677 lines (631 loc) · 37.9 KB
/
methods.html
File metadata and controls
677 lines (631 loc) · 37.9 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
<!DOCTYPE html>
<html lang="en">
<head>
<title>openMetaAnalysis: methods</title>
<meta name="author" content="openMetaAnalysis Contributors">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
<!-- jquery -->
<script src="//ajax.googleapis.com/ajax/libs/jquery/2.1.1/jquery.js"></script>
<script src="/openmetaanalysis.js"></script>
<script>
$(document).ready(function(){
//button handlers
$(".type2").on("click", function(){
if ($("#iqr").is(':checked')){
$("#subjects").css("background-color",lightgray);
}
else{
$("#subjects").css("background-color",white);
}
});
$("#bttncalculate").on("click", function(e){
e.preventDefault();
var subjects = parseFloat($("#denom1").val()) + parseFloat($("#denom2").val()) ;
if ($("#counts").is(':checked')){
weightedvalue = 100*((parseFloat($("#numer1").val()) + parseFloat($("#numer2").val()))/subjects);
//weightedvalue = (($("#numer1").val()*$("#denom1").val() + $("#numer2").val()*$("#denom2").val())/subjects);
weightedvalue = weightedvalue.toFixed(2).toString() + "%";
}
if ($("#percents").is(':checked')){
weightedvalue = (($("#numer1").val()*$("#denom1").val() + $("#numer2").val()*$("#denom2").val())/subjects);
weightedvalue = weightedvalue.toFixed(2).toString() + "%";
}
if ($("#means").is(':checked')){
weightedvalue = (($("#numer1").val()*$("#denom1").val() + $("#numer2").val()*$("#denom2").val())/subjects);
weightedvalue = weightedvalue.toFixed(2).toString();
}
$("#answer").html(weightedvalue);
});
$("#bttncalculate2").on("click", function(e){
e.preventDefault();
var width = Math.abs(parseFloat($("#upper").val()) - parseFloat($("#lower").val()) );
var se = 0;
var sd = 0;
var variance = 0;
var subjects = parseFloat($("#subjects").val());
if ($("#ci").is(':checked')){
se = width/2/1.96 //Chinn, 2000 PMID 11113947
sd = se * Math.sqrt(subjects) //Varies with sample size
}
if ($("#iqr").is(':checked')){
sd = width/1.35;
}
if ($("#range").is(':checked')){
sd = width/4;
}
variance = Math.pow(sd, 2);
if ($("#median").val() != ''){
median = parseFloat($("#median").val())
if ($("#median").val() != '' && $("#range").is(':checked')){
// Hozo, 2005 PMID 15840177: equation 4
var mean = (parseFloat($("#lower").val()) + 2*median + parseFloat($("#upper").val()) )/4 + (parseFloat($("#lower").val()) - 2*median + parseFloat($("#upper").val()) )/(4*subjects)
mean = mean.toFixed(2).toString();
$("#answer_mean").html(mean);
}
else if ($("#median").val() != '' && $("#iqr").is(':checked')){
// Wan, 2014 PMID 25524443: equation 14
var mean = (parseFloat($("#lower").val()) + median + parseFloat($("#upper").val()) )/3
mean = mean.toFixed(2).toString();
$("#answer_mean").html(mean);
}
else {
alert('Selecte Range and enter values for the range in order to convert the median to a mean.')
}
}
se = se.toFixed(2).toString();
$("#answer_se").html(se);
sd = sd.toFixed(2).toString();
$("#answer_sd").html(sd);
variance = variance.toFixed(2).toString();
$("#answer_variance").html(variance);
});
$("#bttncalculate3").on("click", function(e){
//Continous to binary conversion
e.preventDefault();
var width = Math.abs(parseFloat($("#upper3").val()) - parseFloat($("#lower3").val()) );
var or = 1;
var se = 0;
var subjects = parseFloat($("#subjects3").val());
se = width/2/1.96 //Chinn, 2000 PMID 11113947
sd = se * Math.sqrt(subjects) //Varies with sample size
smd = parseFloat($("#meandiff3").val()) / sd;
or = Math.exp(1.81 * smd)
sebin = Math.exp(1.81 * se)
$("#answer3").html(or.toFixed(2).toString() + "(" + (or-1.96*sebin).toFixed(2).toString() + " to " + (or+1.96*sebin).toFixed(2).toString() + ")");
});
});
</script>
<style>
.wrapper
{
width: 860px;
margin-top: 0px;
margin-right: auto;
margin-bottom: 0px;
margin-left: auto;
}
table, td, th{
padding:10px;
border: 1px solid black;
border-collapse: collapse;
}
/* Menu from http://codepen.io/philhoyt/pen/ujHzd */
#primary_nav_wrap
{
margin-top:15px;
}
#primary_nav_wrap ul
{
list-style:none;
position:relative;
float:left;
margin:0;
padding:0
}
#primary_nav_wrap ul a
{
display:block;
background-color:#6DC6E7;
color:#FFFFFF;
text-decoration:none;
font-weight:bold;
//font-weight:700;
//font-size:12px;
line-height:32px;
padding:0 15px;
//font-family:"HelveticaNeue","Helvetica Neue",Helvetica,Arial,sans-serif
}
#primary_nav_wrap ul li
{
position:relative;
float:left;
margin:0;
padding:0
}
#primary_nav_wrap ul li.current-menu-item
{
background:#0022B4;
color:#FFFFFF;
}
#primary_nav_wrap ul a:hover
{
background:#0022B4;
color:#FFFFFF;
}
#primary_nav_wrap ul ul
{
display:none;
position:absolute;
top:100%;
left:0;
background-color:#6DC6E7;
color:#FFFFFF;
padding:0
}
#primary_nav_wrap ul ul li
{
float:none;
width:200px
}
#primary_nav_wrap ul ul a
{
line-height:120%;
padding:10px 15px;
}
#primary_nav_wrap ul ul ul
{
top:0;
left:100%
}
#primary_nav_wrap ul li:hover > ul
{
display:block
}
</style>
</head>
<body>
<!--
http://jquery-csv.googlecode.com/git/examples/basic-usage.html
http://jquery-csv.googlecode.com/git/examples/flot.html -> seems best and most stable without reticulating spines error
http://code.google.com/p/jquery-csv/wiki/API -> instructions
-->
<div class="wrapper">
<div class="page-header">
<div style="color:#0022B4">
<h1>openMetaAnalysis: <span style="font-size:50%">collaborative and continuous</span></h1>
</div>
<nav id="primary_nav_wrap">
<ul id="menu">
<li class="listitem"><a href="http://openmetaanalysis.github.io/" class="menuitem">Home</a></li>
<li class="listitem"><a href="#" class="menuitem">Editors</a>
<ul>
<li><a href="opencpu-launch.htm?https://openmetaanalysis.ocpu.io/home/www/editor.html" class="menuitem">Intervention</a></li>
<li><a href="opencpu-launch.htm?https://openmetaanalysis.ocpu.io/home/www/editor.html#instruct" class="menuitem">Intervention (instructional)</a></li>
<li><a href="opencpu-launch.htm?https://openmetaanalysis.ocpu.io/home/www/editor-dx.html" class="menuitem">Diagnosis</a></li>
<li class="listitem"><a href="https://openmetaanalysis.ocpu.io/home/www/arrows.html" class="menuitem">Arrows Plot</a></li>
</ul>
</li>
<li class="listitem"><a href="http://openmetaanalysis.github.io/methods.html" class="menuitem">Methods</a></li>
<li class="listitem"><a href="http://openmetaanalysis.github.io/new.html" class="menuitem">Create a new review</a></li>
<li class="listitem"><a href="http://openmetaanalysis.github.io/update.html" class="menuitem">Improve/update an existing review</a></li>
</ul>
</nav>
</div>
<div style="clear:both"> </div>
<h1 style="color:#0022B4">Methods</h1>
<div>Methods are adapted from:</div>
<ul>
<li><a href="https://training.cochrane.org/handbook/current">Cochrane Handbook</a>
<ul
<li><a href="https://community.cochrane.org/mecir-manual">Methodological Expectations of Cochrane Intervention Reviews (MECIR)</a></li>
</ul></li>
<li><a href="http://www.gradeworkinggroup.org/">GRADE Working Group</a>
<ul
<li>openMetaAnalysis, similar to meta-analysis in general, focuses on the intial 4 - 6 steps of the GRADE process (see Box 1 at <a href="http://www.ncbi.nlm.nih.gov/pmc/articles/PMC428525/">BMJ, 2006</a>).
<li><a href="https://gdt.gradepro.org/app/handbook/handbook.html">Grade Handbook</a></li>
</ul></li>
<li><a href="http://www.bris.ac.uk/quadas/quadas-2/">QUADAS-2</a> (for meta-analyses of diagnostic tests)</li>
<li><a href="http://www.prisma-statement.org/">PRISMA</a> Statement (PMID: <a href="http://pubmed.gov/19622512">19622512</a>) and Checklist. PRISMA items 7 - 11 are substantially affected by using open and continuous updating.</li>
</ul>
<div>Contents:</div>
<ul>
<li><a href="#study_selection">Files needed</a></li>
<li><a href="#study_selection">Selecton of studies</a></li>
<li><a href="#searching">Searching for studies</a></li>
<li><a href="#data_abstraction">Data abstraction</a> (including data imputing)</li>
<li><a href="#quality_of_studies">Assessing quality of individual studies</a></li>
<ul>
<li><a href="#summary_judgment_for_one_study">Creating a summary judgment for each study</a></li>
</ul>
<li><a href="#analysis">Statistical analysis</a>
<ul><li><a href="#smd">Standardized mean difference (SMD) to pool results of difficult and variably measured outcomes</a></li></ul>
</li>
<li><a href="#quality_of_group_of_studies">Assessing quality across a group of studies</a>. Risk of Bias (RoB) Tables</li>
<li><a href="#sof">Summary of Findings</a></li>
<li><a href="#reconciliation_tables">Reconciliation tables</a></li>
</ul>
<div>
<a name="files_needed"/>
<span style="font-weight:bold">Download examples files, then replace with your content</span>
<span style="font-weight:bold;font-size:130%">Files needed</span>
</div>
<ul>
<li>Reconcilation tables: <a href="https://openmetaanalysis.github.io/reconciliation">https://openmetaanalysis.github.io/reconciliation</a></li>
</ul>
<p><a href="https://github.com/openMetaAnalysis/Template-a-full-living-systematic-review-of-interventions/tree/master/files">Other downloads</a>:</p>
<ul>
<li>Folder 'Data': Outcomes spreadsheets</li>
<li>Folder 'Study-details': Risk of bias tables</li>
<li>Folder 'Study-details': PICO tables</li>
</ul>
<div>
<a name="study_selection"/>
<span style="font-weight:bold;font-size:130%">Selecton of studies</span>
<span style="font-weight:bold"> (PRISMA Item 6)</span>
</div>
<p>Meta-analysis of randomized controlled trials:</p>
<ul>
<li>Random allocation</li>
</ul>
<p>Meta-analysis of diagnostic test accuracy:</p>
<ul>
<li>Cross sectional or cohort design (not case controls)</li>
<li>Random or consecutive selection of patients</li>
<li>All patients received a reference standard</li>
</ul>
<div>More strict criteria for inclusion may be based on the <a href="https://acpjc.acponline.org/shared/purpose_and_procedure.htm">criteria by American College of Physicians Journal Club (ACPJC)</a>.</div>
<div> </div>
<div>We encourage reviews to have a Table of Reconciliation of Studies (<a href="https://github.com/openmetaanalysis/Early-goal-directed-therapy-for-septic-shock/tree/master/reconciliation%20of%20studies/">example</a>) that tabulates studies included by prior reviews and their status in the living review (Not part of PRISMA checklist). The has been advocated previously (PMID <a href="http://pubmed.gov25551377">25551377</a>).
<div> </div>
<div>
<a name="searching"/>
<span style="font-weight:bold;font-size:130%">Searching for studies</span>
<span style="font-weight:bold"> (PRISMA Items 7,8)</span>
</div>
<p>1. Start with making a reconciliation table of all studies included in at least one recent meta-analysis. (PMID <a href="https://pubmed.gov/27136216">27136216</a>)</p>
<ul>
<li>Identify up to 3-4 recent meta-analyses; place these in the columns a a table</li>
<li>List each trial in the rows of a spreadsheet.</li>
<li>In each cell, note whether a study was included or the reason it was not</li>
<li><a href="https://github.com/openMetaAnalysis/Early-goal-directed-therapy-for-septic-shock/tree/master/reconciliation-tables">Example of reconciliation tables</a></li>
</ul>
<p>2. Search for more recent studies with traditional searching using search terms</p>
<ul>
<li>Search PubMed with a 'Brief Search Strategy'(PMID <a href="http://pubmed.gov/16042789">16042789</a>)</li>
<li>Manual searches of online textbooks</li>
<li>For clinical trials
<ul>
<li>Search of the <a href="http://www.cochranelibrary.com/">Cochrane Central Register of Controlled Trials (CENTRAL)</a></li>
<li>Clinical Trials registries with methods suggested by Glanville (<a href="http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4076126/">PMC4076126</a>):
<ul>
<li><a href="https://clinicaltrials.gov/ct2/search/advanced">ClinicalTrials.gov</a> (for studies with completed results and the search term random*)</li>
<li><a href="http://apps.who.int/trialsearch/AdvSearch.aspx">International Clinical Trials Registry Platform (ICTRP)</a> (WHO)</li>
</ul>
</li>
</ul>
</li>
</ul>
<p>3. Search for more recent studies with <a href="#" class="hastip" id="cited_references">cited reference searches</a></p>
<ul>
<li>Identify the seminal (most highly cited) meta-analysis and original study by using <a href="http://scholar.google.com">Google Scholar</a> or <a href="http://isiknowledge.com">Web of Science</a> to retrieve the number of citations to each study on the reconciliaiton table. This work can be reduced by starting with studies in high impact journals. <a href="https://github.com/openMetaAnalysis/Procalcitonin-guided-antibiotics-for-severe-sepsis-and-septic-shock/issues/18">Example</a></li>
<li>Execute cited references searches to retrieve the number of citations to each study in hand, or better, in your reconciliation table by using:
<ul>
<li><a href="http://pubmed.gov">PubMed's</a> "Cited by" feature below each article's abstract.</li>
<li><a href="http://scholar.google.com">Google Scholar</a></li>
<li>Instructions at <a href="https://sumsearch.org/searching/">SUMSearch</a></li>
</ul>
</li>
</ul>
<p>4. Each time a study is identified at PubMed, look for addiitonal studies using <a href="http://www.ncbi.nlm.nih.gov/books/NBK3827/#pubmedhelp.Finding_articles_rel">PubMed's Find Related Data portlet</a></p>
<p>If you find qualifying studies to an existing repository at openMetaAnalysis</p>
<ul>
<li>Please add the citation to the relevant: 1) data table, 2) PICO table, and 3) risk of bias table.</li>
<li>Update the relevant forest plot</li>
<li><i>Consider publishing the updated meta-analysis in a journal.</i></li>
</ul>
Consider <a href="https://library.medicine.yale.edu/tutorials/1559">https://library.medicine.yale.edu/tutorials/1559</a>
<div>
<a name="data_abstraction"/>
<span style="font-weight:bold;font-size:130%">Data abstraction</span>
<span style="font-weight:bold"> (PRISMA Items 9-11)</span>
</div>
<table style="border-style:none;border: none;">
<tr><td>
<div>Creating the files for PICO and Bias tables</div>
<ul>
<li><a href="http://openmetaanalysis.github.io/make-table-pico.html">http://openmetaanalysis.github.io/make-table-pico.html</a> will help make the content for the PICO tables.</li>
<li><a href="http://openmetaanalysis.github.io/make-table-bias.html">http://openmetaanalysis.github.io/make-table-bias.html</a> will help make the content for the Risk of bias table.</li>
</ul>
<div>Consider using an online collaborative text editor with a colleague to develop the xml files for the PICO and bias tables.</div>
<ul>
<li><a href="https://kobra.io">Kobra</a> is very easy to use. Kobra itself is collaboratively developed with <a href="http://www.firepad.io">Firepad</a>.</li>
</ul>
</td>
<td>
<form>
<table>
<caption>Calculation of population characteristics from 'Table 1' of studies for PICO Table in openMetaAnalysis</caption>
<tr><th>Group 1</th><th>Group 2</th></tr>
<tr><td><input type="text" id="numer1" placeholder="Proportion or mean"/></td><td><input type="text" id="numer2" placeholder="Proportion or mean"/></td></tr>
<tr><td><input type="text" id="denom1" placeholder="Number of subjects"/></td><td><input type="text" id="denom2" placeholder="Number of subjects"/></td></tr>
<tr><td colspan="2" align="center"><input id="counts" name="type" type="radio"><label for="counts">Counts</label> <input id="percents" name="type" type="radio"><label for="percents">Percentages</label> <input id="means" name="type" type="radio"><label for="means">Means</label> </td></tr>
<tr><td colspan="2"><button id="bttncalculate">Calculate weighted value</button> Weighted value: <span id="answer"></span></td></tr>
</table>
</form>
<form>
<table>
<caption>Calculation of standard deviation and error from<br/>confidence interval, range, or interquartile range</caption>
<tr><td><input type="text" id="lower" placeholder="Lower limit"/></td><td><input type="text" id="median" placeholder="Median (optional)"/></td><td><input type="text" id="upper" placeholder="Upper limit"/></td></tr>
<tr><td colspan="3" align="center"><input type="text" id="subjects" placeholder="Number subjects in group" style="width: 200px;" /></td></tr>
<tr><td colspan="3" align="center"><input id="ci" name="type2" class="type2" type="radio"><label for="ci">Confidence interval</label> <input id="iqr" name="type2" class="type2" type="radio"><label for="iqr">Interquartile range</label> <input id="range" name="type2" class="type3" type="radio"><label for="range">Range</label></td></tr>
<tr><td colspan="3"><button id="bttncalculate2">Calculate standard deviation</button> Mean: <span id="answer_mean"></span> SE: <span id="answer_se"></span> SD: <span id="answer_sd"></span> VAR: <span id="answer_variance"></span>
<div>Details from: <a href="http://pubmed.gov/25524443,15840177">mean to median</a>, <a href="http://pubmed.gov/11113947">confidence interval</a>, <a href="http://handbook-5-1.cochrane.org/chapter_7/7_7_3_5_mediansand_interquartile_ranges.htm">interquartile range</a>, and <a href="http://www.biomedcentral.com/1471-2288/5/13">range</a>. See downloadable spreadsheet for calculations by Wan, 2014 PMID <a href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4383202/">25524443</a></div>
</td></tr>
</table>
</form>
<form>
<table>
<caption>Converting continuous measures into binary measures</caption>
<tr><td colspan="2">Many calculators for data conversions are at <a href="https://www.campbellcollaboration.org/calculator/">George Mason University</a>. Equations for conversion <a href="https://training.cochrane.org/handbook/current/chapter-10#section-10-6">are available</a> at the Cochrane handbook-5-1.</tr>
<tr><td colspan="2" align="center"><input onclick="alert('Not implemented')" id="twosamples3" name="type2" class="type2" type="radio"><label for="twosamples3">Two group means</label> <input id="smd3" name="type2" class="type2" type="radio"><label for="smd3">Single difference between group means</label></td></tr>
<tr class="onedifference3"><td><input type="text" id="lower3" placeholder="CI lower limit"/></td><td><div></div><input type="text" id="upper3" placeholder="CI upper limit"/></td></tr>
<tr class="onedifference3"><td colspan="2" align="center"><input type="text" id="subjects3" placeholder="Number subjects in group" style="width: 200px;" />
<div><input type="text" id="meandiff3" placeholder="Mean difference" style="width: 200px;" /></div>
</td></tr>
<tr><td colspan="2"><button id="bttncalculate3">Calculate binary measures</button> Odds ratio (under construction): <span id="answer3"></span>
<div>Details from: Chinn, 2000 (PMID <a href="http://pubmed.gov/11113947">11113947</a>) and da Costa, 2014 (PMID <a href="http://pubmed.gov/23045205">23045205</a>).</div>
</td></tr>
</table>
</form>
</td></tr>
</table>
<a name="SMD"/>
<div><a href="http://handbook-5-1.cochrane.org/chapter_9/9_2_3_2_the_standardized_mean_difference.htm">Standardized mean differences</a> may be abstracted. One proposed interpretration of the SMD is: (PMID <a href="http://pubmed.gov/19565683">19565683</a>)</div>
<ul>
<li>0.2 represents a small effect</li>
<li>0.5 a moderate effect</li>
<li>0.8 a large effect</li>
</ul>
<div>
<a name="quality_of_studies"/>
<span style="font-weight:bold;font-size:130%">Assessing quality of individual studies</span>
<span style="font-weight:bold"> (PRISMA Items 12)</span>
</div>
<div>The Cochrane has release Risk of Bias 2, which has been partially implemented here. Links:</div>
<ul>
<li><a href="https://methods.cochrane.org/bias/resources/rob-2-revised-cochrane-risk-bias-tool-randomized-trials">Cochrane announcement</a></li>
<li><a href="https://sites.google.com/site/riskofbiastool/welcome/rob-2-0-tool">Google Site</a></li>
</ul>
<div><a href="https://robotreviewer.vortext.systems">Robot Reviewer</a> can help you locate text in articles to make determinations of biases.</div>
<ul>
<li>Studies of intervention
<ul>
<li><a href="http://handbook-5-1.cochrane.org/chapter_8/table_8_5_d_criteria_for_judging_risk_of_bias_in_the_risk_of.htm">Individual criteria for judging risk of bias</a> in a trial using the '<a href="http://handbook-5-1.cochrane.org/chapter_8/8_5_the_cochrane_collaborations_tool_for_assessing_risk_of_bias.htm">Cochrane's Risk of bias tool</a>'.(1)
<ul>
<li>Attrition: consider judgment of 'Unclear risk" it no Consort flow diagram</li>
<li><a name="selective.reporting"/>Selective reporting: consider judgment of 'Unclear risk" it no access to original protocol at a online trial registry
<ul>
<li><a href="https://raw.githubusercontent.com/openMetaAnalysis/openMetaAnalysis.github.io/master/images/Trial_registration_finding_initial.mp4">Video of finding a trial's initial registration</a> at ClinicalTrials.gov</li>
<li>Check whether date of first registration was prior to date of first patient randomized</li>
<li>Check whether primary and secondary outcomes planned before randomization match those in publications of results</li>
</ul>
</li>
</ul>
</li>
<li><a href="http://handbook-5-1.cochrane.org/chapter_8/table_8_7_a_possible_approach_for_summary_assessments_of_the.htm">Aggregating individual criteria into a summary judgment about a trial</a> (see column "Within a study")</li>
</ul>
</li>
<li>Studies of diagnostic accuracy
<ul>
<li><a href="http://www.bris.ac.uk/quadas/quadas-2/">Four criteria for judging risk of bias from QUADAS-2</a> (see Risk of bias row in the Table guided by the detailed criteria in the rows above)</li>
</ul>
</li>
<li>Studies of prognosis and risk factors
<ul>
<li>Cochrane's <a href="http://methods.cochrane.org/sites/methods.cochrane.org.prognosis/files/public/uploads/QUIPS tool.pdf">Quality In Prognosis Studies (QUIP) tool</a> (PMID <a href="https://pubmed.gov/23420236">23420236</a>)</li>
</ul>
</li>
<li>Conflict of interest
<ul>
<li><a Href="http://www.icmje.org/recommendations/browse/roles-and-responsibilities/author-responsibilities--conflicts-of-interest.html">Per ICJME</a>, "Authors should avoid entering in to agreements with study sponsors, both for-profit and nonprofit,
that interfere with authors’ access to all of the study’s data or that interfere with their ability to analyze and interpret the data and to prepare and publish manuscripts
independently when and where they choose.". Thus, adding to selective reporting or publication bias if authors without conflicts did not vouch for the data. However, disclosure on conflicts of interest bu authors may not be complete (<a href="http://pubmed.gov/25835490">25835490</a></li>
</ul></li>
</ul>
<a name="summary_judgment_for_one_study"/>Summary judgment is from the <a href="http://handbook-5-1.cochrane.org/chapter_8/table_8_7_a_possible_approach_for_summary_assessments_of_the.htm">Cochrane Handbook, Table 8.7a</a>
<ul>
<li>Low risk of bias: "Low risk of bias for all key domains."</li>
<li>Unclear risk of bias: "Unclear risk of bias for one or more key domains."</li>
<li>High risk of bias: "High risk of bias for one or more key domains."</li>
</ul>
<div>
<a name="analysis"/>
<span style="font-weight:bold;font-size:130%">Statistical analysis</span>
<span style="font-weight:bold"> (PRISMA Items 13 - 16)</span>
</div>
<p>Analyses are done with online at <a href="https://www.opencpu.org/">OpenCPU</a> using <a href="http://cran.r-project.org/">R</a>. Editors are available for <a href="https://openmetaanalysis.ocpu.io/home/www/editor.html">randomized controlled trials</a> and <a href="https://openmetaanalysis.ocpu.io/home/www/editor-dx.html">diagnostic tests accuracy studies</a>.</p>
<ul>
<li>Studies of intervention
<ul>
<li>We use the random effects model with the inverse variance as implemented in the R package <a href="http://cran.r-project.org/web/packages/meta/">meta</a>.</li>
<li>The Knapp-Hartung method adjusts test statistics and confidence intervals to gives wider (more conservative) confidence intervals and is one of the methods suggested by Cornell et al.(PMID <a href="http://pubmed.gov/24727843">24727843</a>)</li>
<li>The continuity correction of Diamond is used.(PMID: <a href="http://pubmed.gov/17679700">17679700</a>)</li>
<li>Summary measures for binary outcomes include odds ratio and relative risk. Measures for continuous outcomes are either mean differences or <a href="http://handbook-5-1.cochrane.org/chapter_12/12_6_interpreting_results_from_continuous_outcomes_including.htm" class="hastip" id="smd_definition">standardized mean differences</a>.</li>
<li>Subgroup analyses and meta-regressions are done as needed with the R package <a href="http://cran.r-project.org/web/packages/meta/">meta</a>. This includes investigating correlation of the control rate with the outcome
<ul><li><a href="http://pubmed.gov/24339017,21908417,15358396,9777687?dopt=abstract">PubMed citations</a></li></ul>
</li>
<li>Heterogeneity is assessed with I<sup>2</sup> (4).
<ul>
<li>I<sup>2</sup> is "the percentage of the variability in <i>effect estimates</i> that is due to heterogeneity rather than sampling error (chance)".</li>
<li>τ<sup>2</sup>, in random effects meta-analysis, is heterogeneity in <i>variance</i> from random-effects meta-analysis (<a href="https://handbook-5-1.cochrane.org/chapter_9/9_5_4_incorporating_heterogeneity_into_random_effects_models.htm">link from Cochrane Handbook</a>).</li>
<li>τ, in random effects meta-analysis, is "is the estimated <i>standard deviation</i> of underlying effects across studies" (<a href="https://handbook-5-1.cochrane.org/chapter_9/9_5_4_incorporating_heterogeneity_into_random_effects_models.htm">link from Cochrane Handbook</a>).</li>
</ul>
</li>
</ul>
</li>
<li>Studies of diagnosis
<ul>
<li>Hierarchial bivariate model (Reitsma, 2005. PMID <a href="http://pubmed.gov/16168343">16168343</a>) as implemented in the R package <a href="http://cran.r-project.org/web/packages/Metatron/">metatron</a> with AUC from the R package <a href="http://cran.r-project.org/web/packages/mada/">mada</a></li>
</ul>
</li>
<a name="smd"/>
<ul>
<li>The standardized mean difference (SMD) may help combine studies that report conceptually compatible metrics, but vary in the exact method of measurement yielding conflicting numerical results. Many statistical measures used in healthcare can be converted to the SMD (https://training.cochrane.org/handbook/current/chapter-10#section-10-6">Cochrane Handbook</a>) and then pooled (<a href="https://www.campbellcollaboration.org/calculator/">online calculators at the Campbell Collaboration</a>, <a href="https://cran.r-project.org/web/packages/esc/">R package calculators</a>). The SMD can be interpreted as 0.2 represents a small effect; 0.5 a moderate effect; 0.8 a large effect (PMID <a href="http://pubmed.gov/19565683">19565683</a>). As the SMD may not be familiar to readers and also does not give a result that can be translated into clinical impact, after the SMD is calculated and a statistically significant result is found, you can identify a study whose result is closest to the pooled SMD and then describe that study. For example, in the <a href="https://github.com/openMetaAnalysis/Adult_ADHD_Diagnosis/blob/master/README.md">diagnosis of adult ADHD</a>, the <a href="https://pubmed.ncbi.nlm.nih.gov/32285644/">Brevick study</a> had the closest SMD to the pooled SMD of the two studies of the WURS. Thus you could describe the Brevick results in clinical terms using the aROC that they reported and ADHD experts may understand.</li>
</ul>
</ul>
<div>
<a name="quality_of_group_of_studies"/>
<a id="user-content-assessing-quality-of-a-group-of-studies" class="anchor" href="#assessing-quality-of-a-group-of-studies" aria-hidden="true"><span class="octicon octicon-link"></span></a>
<span style="font-weight:bold;font-size:130%">Assessing quality across a group of studies</span>
<span style="font-weight:bold"> (PRISMA Items 15)</span>
</div>
<p><a href="http://handbook-5-1.cochrane.org/chapter_12/table_12_2_b_factors_that_may_decrease_the_quality_level_of_a.htm">Factors</a> developed by the <a href="http://www.gradeworkinggroup.org/">GRADE Working Group</a> are below for assessing a group of studies in a meta-analysis (PMID <a href="http://pubmed.gov/22542023">22542023</a>). (2) Specific criteria for each factor are are based on those used by the <a href="http://back.cochrane.org/">Cochrane Back Group</a> with modifications noted below.(PMID: <a href="http://pubmed.gov/23362516">23362516</a>)</p>
<table>
<tbody>
<tr style="background-color:lightgray">
<th>Factors</th>
<th>Criteria</th>
</tr>
<tr>
<td>Limitations in the design and implementation of available studies</td>
<td>
Cochrane Back Group (<a href="http://pubmed.gov/23362516">23362516</a>)
<ul>
<li>Serious risk of bias: More than 25% of participants from studies with low methodological quality as measured by the Cochrane's (interventions) or QUADAS-2 (diagnostic tests) Risk of bias tool (see above)</li>
<li>Very serious risk of bias: More than 50% of participants from studies with low methodological quality as measured by the Cochrane's (interventions) or QUADAS-2 (diagnostic tests) Risk of bias tool"</li>
</ul>
Alternative approach: <a href="http://handbook-5-1.cochrane.org/chapter_8/table_8_7_a_possible_approach_for_summary_assessments_of_the.htm">Cochrane Handbook 5.1; Table 8.7</a>
Alternative approach: <a href="https://training.cochrane.org/handbook/version-6/chapter-14-draft">Cochrane Handbook 6.0; Table 14.2.b</a>
<ul>
<li>Low risk of bias: "Most information is from results at low risk of bias."</li>
<li>Some concerns: "Most information is from results at low risk of biasor with some concerns."</li>
<li>High risk of bias: "The proportion of information from results at high risk of bias is sufficient to affect the interpretation of results."</li>
</ul>
</td>
</tr>
<tr>
<td>Indirectness</td>
<td><a href="http://handbook-5-1.cochrane.org/index.htm#chapter_8/8_7_summary_assessments_of_risk_of_bias.htm">Cochrane Handbook 5.0</a></td>
</tr>
<tr>
<td>Heterogeneity or inconsistency of results using I<sup>2</sup><br/>(defined as "percentage of total variation across studies that is due to heterogeneity rather than chance." Higgins, 2003 PMID <a href="http://pubmed.gov/12958120">12958120</a>)</td>
<td>From <a href="https://training.cochrane.org/handbook/current/chapter-10#section-10-10-2">Identifying and measuring heterogeneity</a>),
a 'rough guide' to interpretation is:<br/>
• >0% to 40%: might not be important<br/>
• >30% to 60%: may represent moderate heterogeneity<br/>
• 50% to 90%: may represent substantial heterogeneity<br/>
• 75% to 100%: considerable heterogeneity
</td>
</tr>
<tr>
<td>Imprecision of results<br>(modified from the Cochrane Back Group)</td>
<td>• Serious imprecision: Fewer than 2000 participants for each outcome (PMID: <a href="http://pubmed.gov/11158556">11158556</a>) or confidence intervals that include clinically unimportant outcomes<br>• Very serious imprecision: Fewer than 300 participants for each outcome.(PMID: <a href="http://pubmed.gov/23362516">23362516</a>)
<div>An alternative approach to determining imprecision is to use the "optimal information size" (PMID: <a href="http://pubmed.gov/21839614">21839614</a>) with an online calculator (<a href="http://www.stat.ubc.ca/~rollin/stats/ssize/b2.html">http://www.stat.ubc.ca/~rollin/stats/ssize/b2.html</a>).</div>
</td>
</tr>
<tr>
<td>Probability of publication bias</td>
<td>This area of meta-analytic practice is evolving and presently only addresses studies of interventions. See discussion at <a href="http://handbook-5-1.cochrane.org/chapter_10/10_4_5_summary.htm">http://handbook-5-1.cochrane.org/chapter_10/10_4_5_summary.htm</a>.
<ul>
<li>If more than 10 studies are present, test for the small study effect with the Egger test for continuous outcomes or the Rucker test for binary outcomes (<a href="http://cran.r-project.org/web/packages/metasens/">CRAN</a> and PMIDs: <a href="http://pubmed.gov/17592831,19836925">17592831,19836925</a>).</li>
<li>When less than 10 studies are present, study size of less than 50 or 1000 patients total (PMID: <a href="http://pubmed.gov/23616031">23616031</a>) or 100 per arm (PMID: <a href="http://pubmed.gov/20639294">20639294</a>) in most of all available studies may suggest small study effect.</li>
<li>Selective reporting risk if trial not registered. Selective reporting is common (<a href="http://pubmed.gov/23407296,26287998,19724045">23407296,26287998,19724045</a>). Trial registration, may lead to less favorable conclusions.PMID: <a href="http://pubmed.gov/26244868,22214754">26244868,22214754</a></li>
</ul>
</td>
</tr>
<tr>
<td>Rating up the evidence</td>
<td>GRADE recommends rating up the evidence if one of the following exist (PMID: <a href="http://pubmed.gov/21802902">21802902</a>). Note similarity to Bradford-Hill criteria (PMID <a href="http://pubmed.gov/14283879">14283879</a>):
<ul>
<li>"GRADE suggests considering rating up quality of evidence one level when methodologically rigorous observational studies show at least a two-fold reduction or increase in risk, and rating up two levels for at least a five-fold reduction or increase in risk"</li>
<li>Dose-response gradient is present</li>
<li>"All plausible confounders or biases would decrease an apparent treatment effect"</li>
<li>Rapidity of the response</li>
</ul>
</td>
</tr>
</tbody></table>
<div> </div>
<div>
<a name="sof"/>
<span style="font-weight:bold;font-size:130%">Summary of Findings</span>
<span style="font-weight:bold"> (Not part of PRISMA checklist)</span>
</div>
<div> </div>
<div>If pooled studies show significant benefit</div>
<ul><li>Summary of Finding Tables (SoF) are constructed as detailed in the Cochrane Handbook (<a href="http://handbook-5-1.cochrane.org/chapter_11/11_5_summary_of_findings_tables.htm">Chapter 11</a>) and by Guyatt et al (PMID: <a href="http://pubmed.gov/21195583">21195583</a>). SoF are currently produced with <a href="http://tech.cochrane.org/revman/gradepro">GRADEpro</a>.</li></ul>
<a name="judicial"/>
<div>If pooled studies <i>do not</i> show significant benefit</div>
<ul><li>Results are summarized based on p-values using judical analogies developed by Diamond (PMID <a href="http://pubmed.gov/19667308">19667308</a>).</ul>
<table class="tg">
<tr style="background-color:lightgray">
<th>P values</th>
<th>Judicial analogy</th>
</tr>
<tr>
<td>P < 0.05 in all studies</td>
<td>“Beyond a reasonable doubt”</td>
</tr>
<tr>
<td>P < 0.05 in some studies</td>
<td>“Clear and convincing evidence”</td>
</tr>
<tr>
<td>P < 0.5 in all studies</td>
<td>“Preponderance of the evidence”</td>
</tr>
<tr>
<td>P < 0.5 in some studies</td>
<td>“Reasonable suspicion”</td>
</tr>
<tr>
<td>P < 0.5 in no studies</td>
<td>“Insufficient evidence”</td>
</tr>
<tr>
<td colspan=2>Based on Diamond, 2009 (PMID <a href="http://pubmed.gov/19667308">19667308</a>)</td>
</tr>
</table>
<div> </div>
<div>
<a name="reconciliation_tables"/>
<span style="font-weight:bold;font-size:130%">Reconciliation of Conclusions with prior meta-analyses</span>
<span style="font-weight:bold"> (Not part of PRISMA checklist)</span>
</div>
<div>The has been advocated (PMID <a href="http://pubmed.gov25551377">25551377</a>) and implemented (<a href="http://pubmed.gov/25796995">25796995</a>) previously.</div>
<div><a href="https://github.com/openMetaAnalysis/Early-goal-directed-therapy-for-septic-shock/blob/master/Reconciliation%20of%20conclusions.csv">Example table</a>.</div>
<div> </div>
<h3>References</h3>
<div>
<ol class="task-list">
<li><a href="http://handbook-5-1.cochrane.org/chapter_8/8_5_the_cochrane_collaborations_tool_for_assessing_risk_of_bias.htm">Risk of bias tool</a>. In: Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. Available from <a href="http://handbook-5-1.cochrane.org">http://handbook-5-1.cochrane.org</a></li>
<li><a href="http://handbook-5-1.cochrane.org/chapter_12/12_2_1_the_grade_approach.htm">The GRADE approach</a>. In: Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. Available from <a href="http://handbook-5-1.cochrane.org">http://handbook-5-1.cochrane.org</a></li>
<li>Schünemann H, Brożek J, Guyatt G, Oxman A (editors). <a href="http://www.guidelinedevelopment.org/handbook/">GRADE Handbook</a>. [updated October 2013]. Available from <a href="http://www.guidelinedevelopment.org/handbook/">http://www.guidelinedevelopment.org/handbook/</a></li>
<li><a href="http://handbook-5-1.cochrane.org/chapter_9/9_5_2_identifying_and_measuring_heterogeneity.htm">Identifying and measuring heterogeneity</a>. In: Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. Available from <a href="http://handbook-5-1.cochrane.org">http://handbook-5-1.cochrane.org</a></li>
</ol>
</div>
<div>Click here to see the <a href="https://github.com/openMetaAnalysis/home/pulse">revision history</a> of the source code.</div>
<div> </div>
<script>
//For gh-pages
//Page history and edit
var pagename = location.pathname.split('/').slice(-1);
if (pagename.toString().length < 1){pagename = "index.html"}
document.write("<div style='text-align:center'><a href='https://github.com/openMetaAnalysis/openMetaAnalysis.github.io/blob/master/" + pagename + "'>Edit this page</a> - <a href='https://github.com/openMetaAnalysis/openMetaAnalysis.github.io/commits/master/" + pagename + "'>Page history</a></div>")
</script>
</div>
<script src="javascripts/scale.fix.js"></script>
<script type="text/javascript">
var gaJsHost = (("https:" == document.location.protocol) ? "https://ssl." : "http://www.");
document.write(unescape("%3Cscript src='" + gaJsHost + "google-analytics.com/ga.js' type='text/javascript'%3E%3C/script%3E"));
</script>
<script type="text/javascript">
try {
var pageTracker = _gat._getTracker("UA-56740469-1");
pageTracker._trackPageview();
} catch(err) {}
</script>
<div id="tip" style="display:none"></div>
</body>
</html>