Concealed Allocation



Concealed Allocation [pic] [pic] [pic]

Allocation, concealed. Part of the mnemonic for judging the quality of therapy RCTs: ABCDFIX.

Concealed allocation is a procedure implemented in a randomized control trial where the individuals screening and separating the candidates into two (or more) arms of a study are blinded.

This is a consideration beyond blinding the practitioner delivering the care or the patients receiving the care.

Why does it matter?

Even if there is a randomization process in place, it matters how it is actually implemented.

To be effective, the process must ensure that the investigators (or providers or subjects) CANNOT influence the group each person ends up in. Thus the allocation into the groups must be “concealed.” Otherwise, the researcher, provider delivering care, or the subjects themselves may consciously or unconsciously manipulate who gets allocated to which arm of the study—thusly defeating the benefits of randomization.

Does it REALLY matter that much?

It can. Trials with unconcealed allocation, compared to trials on the same interventions where allocation was concealed, consistently overestimated the benefit of a treatment by as much as 40% (as measured by odds ratios)! (Schultz 1995, Schultz 2002, Pidal 2007, Moher 1998) When trials with inadequate concealment were dropped from a random sampling of 38 Cochrane

systematic reviews, only 48 of 70 conclusions regarding the therapy under review held up(Pildal 2007).

In a meta-analysis of multiple trials studying whether mammography was useful as a screening procedure for breast cancer, a significant difference was found between those studies where allocation was concealed compared to those that were not.

Mammography

Mammography was found to be useful in studies where allocation was NOT concealed. But there turned out to be a number of differences in the screened woman versus the unscreened control group of women. The screened group were in a higher socioeconomic class, were more educated, and were slightly younger. Did any of these characteristics make a difference? Perhaps so. In the studies where allocation was concealed, these disparities disappeared resulting in groups that were more evenly matched. When this happened, the benefits disappeared. In fact, there was not only no effect on mortality but there were 20% more mastectomies! For women between the age of 40 and 50, there continues to be controversy regarding the merits of screening all women in that age group with mammography.

Surfactant and premies

An RCT in the early 1990s evaluated the benefit of artificial surfactant for premature infants in the neonatal intensive care unit. Whether a premature infant was administered surfactant or not depended on what was contained in a set of randomized sealed envelopes. Although the envelops were sealed, they were not reported to be opaque. It was possible for physicians to hold the envelopes up to a light and determine whether or not the next baby was going to get surfactant.

The study would be significantly flawed if the care givers were to have a bias toward giving surfactant (administering surfactant was, after all, the norm and what they had been taught in medical school). By backlighting the study envelopes, they could tell whether the next neonate was to get surfactant or not. They then could choose to enroll neonates whose status was marginal and who were not likely to survive whenever a placebo envelop randomly came up. They could also make sure that those with a good chance for survival got the surfactant. If a placebo envelop was up next for these infants, they simply would not enroll the neonate in the study, freeing them to administer surfactant anyway.

As a result, sicker infants would have been selectively enrolled into the placebo group, healthier infants in the surfactant group. This selectivity could result in greatly overestimating the benefit of surfactant. This study was published in the medical literature as a randomized, double-blind controlled trial. Rumors surfaced later that a number of the care givers breached the protocol. Since the authors did not specify that the envelopes were opaque, readers would not know for certain that allocation was concealed.

Bottom line: When allocation is not concealed, it doesn’t mean that the results are not valid, but it means we don’t know. Because it can potentially introduce a lot of bias, most researchers now would consider the study significantly flawed.

How is concealment done?

Allocation occurs as one step of the research process before the trial really begins. First a call goes out for potential subjects (sampling). Then this pool of respondents is culled down into a smaller pool based on certain inclusion and exclusion criteria. At that point the subjects can be divided and allocated into the different research arms.

[pic]

Note: the method of randomization is one consideration (e.g., computer, drawing names out of a hat). Actually placing “allocated” subjects into each research arm is a separate consideration. Does the person distributing the subjects according to the randomization scheme know which groups each subject is ending up in? If this information is not concealed, the allocator could potentially alter the randomization process.

On the other hand allocation concealment can occur even in an otherwise unblinded study. Let’s look at an example:

Knee DJD

A study on knee surgery (Moseley 2002) illustrates how a study can be “unblind” in the traditional sense but yet the allocation can be concealed.. The study evaluated patients with degenerative joint disease of the knee who were randomized to receive physical therapy or surgery. The patients and the researcher were not told which treatment the patients would get (allocation was therefore concealed) until after they were enrolled into the study. If they had known what they were going to get, either surgery or physical therapy, the patients may have elected not to be enrolled, and, therefore, the generalizability of the study might have been open to question. Once they were in the study, though, both the patients and the treating physician or physical therapist were certainly aware of the treatment the patient was receiving.

Methods of Concealment

The Good

Allocation concealment is best achieved by using a centralized off-site computer allocation process. This method is usually used for large multicentre trials.

For smaller trials, use of an independent person and a sealed, opaque envelope system gives a satisfactory result.

The Questionable

The provider or researcher handing out sealed envelopes, rather than an independent third party, is not as good because the allocation may not be as well concealed. Or if the envelops are not reported as both sealed and opaque.

The Ugly

Methods such as allocating alternate subjects, allocating subjects to groups on alternate days as they first come in for treatment, or selecting subjects from databases, do not conceal allocation and are not considered randomized controlled trials.

The Bad

The study does not conceal allocation at all, does not say one way or the other, or it’s just too hard to tell whether they did or not.

Where do I find this information?

Look at the METHODS section of the article you are reading. You can usually tell in about 30 seconds. Even if the study is labeled as “double blind,” the allocation step might not have been concealed.

From a study on ankle sprains:

“Randomization was stratified by centre, and administered independently by a central telephone randomization centre (Birmingham Cancer Trials Service), ensuring allocation concealment.”

(Lamb 2009)

Comment: This doesn’t really tell us how the randomization was done, but the fact that an independent center did the allocation and that they stated that allocation was concealed gives us some assurance. Verdict: Probably concealed allocation.

_________________________________________

From a study on kava and depression:

“Allocation to treatment groups was undertaken by an independent researcher, using sets of random permutations, in order to ensure approximately equal numbers of participants in each group. Randomization details were provided in an opaque envelope containing the allocation (labeled by group number to retain blinding).” (Sarris 2009)

Comment: “Using sets of random permutations” tells us how the randomization was done. The fact that implementation of this randomization process was done by an independent researcher in opaque envelops is good. Better if it had been explicitly stated that they were sealed. Verdict: Concealed allocation was probably adequate (but sealed envelopes would have added more certainty).

_______________________________________

From a study comparing shock wave therapy and exercise for Achilles tendinopathy:

“A computerized random-number generator was used to formulate an allocation schedule. Block randomization (permuted blocks of 3) was implemented. A medical assistant allocated interventions via opaque sealed envelopes marked according to the allocation schedule. The medical assistant was unaware of the size of the blocks.” (Rompe 2007)

Comment: A “computerized random-number generator” tells us how the randomization was done. The fact that medical assistant used opaque, sealed envelops and was unaware of the size of blocks is encouraging. Verdict: Concealed allocation was done!

.

_______________________________________

From a study comparing hydroptherapy and land based exercise of OA of the knee.

“Randomized allocation into either a land-based exercise group or a water-based exercise group was done by drawing lots.” (Silva 2008)

Comment: “drawing lots” sound like an acceptable way to randomized. What’s hard to tell is whether the practitioners/researchers were blind to the distribution and notification process. Verdict: Probably concealed allocation, but less sure than any of the examples above.

________________________________________________________________________

TRUE & FALSE QUIZ

1. It is possible to have concealed allocation even if the study is not double blind.

2. The following study appears to have concealed allocation: A newspaper ad is run for patients with migraine headaches. A researcher, blinded to the allocation into treatment and placebo groups, sorts out those subjects who are excluded from the study from those who successfully meet the inclusion criteria.

3. The following process would qualify as concealed allocation: red and white poker chips are randomized by shaking a bag, and then a chip is randomly handed to each sequential patient with white chips getting Co Enzyme Q and red chips getting an identical placebo.

4. The following would qualify as concealed allocation: In a study using therapeutic ultrasound, the practitioner is blinded to whether the machine is working at a therapeutic dose or not.

5. The following would qualify as concealed allocation: both the patient and doctor delivering the care are blinded to the therapy.

________________________________________________________________________

R LeFebvre, DC

Reviewed by M Haas, DC

1/26/11

References

Glasziou P, Del Mar C, Salisbury J. Evidence-Based Practice Workbook Second Edition 2007.

Lamb SE, Marsh JL, Hutton JL, Nakash R, Cooke MW. Mechanical supports for acute, severe ankle sprain: a pragmatic, multicentre, randomized controlled trial. The Lancet 2009;373:575-81.

Moher D, Ba’Pham AJ, Cook, DJ et al. Does quality of reports of randomized trials affect estimates of intervention efficacy reported in meta-analysis? The Lancet 1998;352(9128):609-13

Moseley JB, O'Malley K, Petersen NJ, et al. A controlled trial of arthroscopic surgery for osteoarthritis of the knee. N Engl J Med 2002; 347:81-8.

Pidal J,Hrobjartsson A, et al. Impact of allocation concealment on conclusions drawn from meta-analyses of randomized trials. Int J Epidemiol 2007;36:847-857

Rompe JD, Nafe B, Furia JP, Maffulli N. Eccentric loading, shock-wave treatment, or a wait-and-see policy for tendinopathy of the main body of Tendo Achillis: a randomized controlled trial. The American Journal of Sports Medicine 2007;35(3):374-83.

Sarris J, Kavanagh DJ, Byrne G, Bone KM, Adams J, Deed G. The kava anxiety depression spectrum study (KADSS): a randomized, placebo-controlled crossover trial using an aqueous extract of piper methysticum. Psychopharmacology 2009; 205:399-407

Schultz KF, Chalmers I, Hayers RJ, et al. Empirical evidence of bias JAMA 1995;273:408-12

Schulz KF, Grimes DA. Allocation concealment in randomized trials: defending against deciphering. The Lancet 2002;359(9306):614-18.

Silva LE, Valim V, Pessanha APC, Oliveira LM, Myamoto S, Jones A, Natour J. Hydrotherapy versus conventional land-based exercise for the management of patients with osteoarthritis of the knee: a randomized clinical trial. Physical Therapy 2008;88(1):12-21

Concealed Allocation TRUE & FALSE QUIZ answers

1. It is possible to have concealed allocation even if the study is not double blind.

True. Absolutely! In a study looking at manipulation or surgery, it is common that neither the doctors delivering the care nor the patients are blinded to what is happening. But the researcher allocating which patients get surgery or manipulation can be concealed (blinded).

2. The following study appears to have concealed allocation: A newspaper ad is run for patients with migraine headaches. A researcher, blinded to the allocation into treatment and placebo groups, sorts out those subjects who are excluded from the study from those who successfully meet the inclusion criteria.

False. It is true that the researcher is blinded—but it is at the wrong step! He is blinded between the sampling stage of the project and the step where the pool of acceptable candidates is finalized. For allocation to be concealed it has to happen at the next phase-- when that pool of subjects is randomized into the different arms of the study.

3. The following process would qualify as concealed allocation: red and white poker chips are randomized by shaking a bag, and then a chip is randomly handed to each sequential patient with white chips getting Co Enzyme Q and red chips getting an identical placebo.

False. Mixing poker chips in a bag is probably an acceptable method of randomization. But we don’t know whether the researcher handing out the poker chips knows where the subjects are going to end up. If allocation is not concealed, the researcher could pull a chip out and then decide whom to give it to depending on what the chip meant.

4. The following would qualify as concealed allocation: In a study using therapeutic ultrasound, the practitioner is blinded to whether the machine is working at a therapeutic dose or not.

False. There is blinding here, but not of the person deciding which patients were allocated to the true treatment and which got the sham. We cannot assume that the allocation step was concealed.

5. The following would qualify as concealed allocation: both the patient and doctor delivering the care are blinded to the therapy.

False. Same reason as above. Allocation concealment is not fundamentally based on blinding the patient or the doctor. Lack of concealed allocation can corrupt the attempt at blinding the patient and doctor, but it is actually based on blinding the allocator.

-----------------------

[pic]

Randomization and concealed (blind) allocation

Subjects who meet the inclusion criteria

Manipulation group

Exercise group

Manipulation + exercise group

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download