Is working memory training the priority for ADHD patients?

by

ADHD Attention Deficit DisorderThere are sev­er­al rea­sons why it is impor­tant to devel­op evi­dence-based ADHD treat­ments in addi­tion to med­ica­tion and behav­ior therapy.

Not all chil­dren ben­e­fit from med­ica­tion, some expe­ri­ence intol­er­a­ble side effects, and many con­tin­ue to strug­gle despite the ben­e­fits pro­vid­ed by medication.

Behav­ior ther­a­py can be dif­fi­cult for par­ents to con­sis­tent­ly imple­ment, and does not gen­er­al­ly reduce behav­ior dif­fi­cul­ties to nor­ma­tive levels.

Fur­ther­more, although both treat­ments can help man­age ADHD symp­toms, they gen­er­al­ly do not induce changes that per­sist after treat­ment ends.

Final­ly, despite numer­ous stud­ies doc­u­ment­ing the short- and inter­me­di­ate term ben­e­fits of med­ica­tion and behav­ior ther­a­py, their impact on chil­dren’s long-term suc­cess remains to be clear­ly documented.

In response to these lim­i­ta­tions, researchers have shown grow­ing inter­est in whether cog­ni­tive train­ing — gen­er­al­ly done via com­put­er — can induce more last­ing changes in chil­dren’s abil­i­ty to focus and attend. One approach that has shown promise in help­ing youth with ADHD, and which is now wide­ly avail­able, is Work­ing Mem­o­ry Training.

Work­ing Mem­o­ry Train­ing is based on find­ings that Work­ing mem­o­ry (WM) — the abil­i­ty to hold and manip­u­late infor­ma­tion in mind for sub­se­quent use — is fre­quent­ly com­pro­mised in youth with ADHD and may con­tribute sig­nif­i­cant­ly to symp­toms of inat­ten­tion. WM deficits also to con­tribute to the aca­d­e­m­ic strug­gles that many chil­dren with ADHD expe­ri­ence. Devel­op­ing an inter­ven­tion to enhance WM in chil­dren with ADHD could thus be extreme­ly helpful.

Sev­er­al pub­lished stud­ies sug­gest that WM train­ing is a promis­ing inter­ven­tion for chil­dren with ADHD. In one, chil­dren with ADHD were ran­dom­ly assigned to high inten­si­ty (HI) or low inten­si­ty (LI) WM train­ing. The HI treat­ment involved per­form­ing com­put­er­ized WM tasks, e.g., remem­ber­ing the sequence in which lights appeared in dif­fer­ent por­tions of a grid, recall­ing a sequence of num­bers in reverse order, where the dif­fi­cul­ty lev­el was reg­u­lar­ly adjust­ed to match the child’s per­for­mance by increas­ing or decreas­ing the items to be recalled. This is called ‘adap­tive’ train­ing because the dif­fi­cul­ty lev­el adapts to match the child’s per­for­mance and chil­dren are con­sis­tent­ly chal­lenged to expand their work­ing mem­o­ry capacity.

In the LI con­di­tion, the tasks were sim­i­lar but the dif­fi­cul­ty remained low through­out, i.e., the num­ber of items did not increase when chil­dren respond­ed cor­rect­ly. For these chil­dren, their work­ing mem­o­ry capac­i­ty was not con­sis­tent­ly chal­lenged and was not expect­ed to grow as a result. This was con­sid­ered the con­trol condition.

Each group trained 30–40 min­utes per day, 5 days per week, for 5 weeks with train­ing super­vised by par­ents. Par­ents were sup­port­ed through week­ly phone calls with a trained coach whose role was to help make sure train­ing was imple­ment­ed as intended.

Results indi­cat­ed that imme­di­ate­ly after treat­ment — as well as 3 months lat­er — chil­dren in the HI group showed improved WM per­for­mance com­pared to LI chil­dren. Fur­ther­more, par­ent reports indi­cat­ed sig­nif­i­cant reduc­tions in ADHD symp­toms, par­tic­u­lar­ly inat­ten­tive symp­toms; these reduc­tions remained evi­dent at 3 months. How­ev­er, no ben­e­fits in ADHD symp­toms were evi­dent in reports pro­vid­ed by chil­dren’s teach­ers. Giv­en the impor­tance of improv­ing atten­tion in the class­room, this was a sig­nif­i­cant limitation.

A sub­se­quent study also used ran­dom assign­ment to HI vs. LI train­ing, and observed the impact on chil­dren’s behav­ior in a con­trolled class­room set­ting. Results indi­cat­ed sig­nif­i­cant reduc­tions in off-task class­room behav­ior among chil­dren with ADHD who received HI train­ing. This par­tial­ly address­es con­cerns about fail­ure to find teacher report­ed ben­e­fits in oth­er stud­ies. Chil­dren also showed gains in non-trained mea­sures of WM.

Results from these stud­ies, along with sev­er­al oth­ers, sug­gest that Work­ing Mem­o­ry Train­ing (the spe­cif­ic train­ing sys­tem used in these stud­ies was Cogmed Work­ing Mem­o­ry Train­ing, i.e., CWMT) yields ben­e­fits in non-trained mea­sures of WM and reduc­tions in par­ent-report inat­ten­tive behav­ior. How­ev­er, no study has found ben­e­fits in teacher report­ed behav­ior and symptoms.

Con­cerns about the evi­dence base for CWMT

A sig­nif­i­cant lim­i­ta­tion in the evi­dence-based for using CWMT to treat youth with ADHD is the absence of teacher report­ed ben­e­fits. In addi­tion, some researchers ques­tion whether the LI train­ing is an ade­quate con­trol con­di­tion. This is because although chil­dren in HI and LI train­ing com­plete the same num­ber of tri­als each ses­sion, the LI train­ing takes less time each ses­sion because it does not become more dif­fi­cult. Thus, the con­di­tions dif­fer in ways oth­er than whether dif­fi­cul­ty lev­el adjusts to match the child’s performance.

Some have also sug­gest­ed that par­ents of LI chil­dren may become aware that their child has been assigned to the con­trol group. If par­ent are not tru­ly ‘blind’ to con­di­tion, it could explain par­ent-report­ed ben­e­fits that have been found. For these rea­sons, some have sug­gest­ed that CWMT should be regard­ed as no more than a ‘pos­si­bly effi­ca­cious’ treat­ment for ADHD and not con­sid­ered a ‘first-line’ treat­ment like med­ica­tion and behav­ior therapy.

Results from 2 recent trials

Two recent­ly pub­lished stud­ies pro­vide impor­tant new data on the effi­ca­cy of CWMT for ADHD.

The first [van Don­gen-Booms­ma et al., (2014). Work­ing mem­o­ry train­ing in young chil­dren with ADHD: A ran­dom­ized con­trolled tri­al] was con­duct­ed with 51 5–7‑year old chil­dren with ADHD in the Nether­lands. Sim­i­lar to the stud­ies sum­ma­rized above, chil­dren were ran­dom­ly assigned to HI vs. LI train­ing. Train­ing con­sist­ed of 25 ses­sions of 15 min­utes 5 days a week for 5 weeks; this is the rec­om­mend­ed train­ing sched­ule for younger chil­dren. Train­ing was con­duct­ed in chil­dren’s home and super­vised by par­ents. Train­ing was con­duct­ed in chil­dren’s home and super­vised by parents.

A cer­ti­fied coach con­tact­ed par­ents each week to eval­u­ate the per­for­mance and moti­va­tion of the child using a stan­dard­ized ques­tion­naire. Nei­ther child, par­ents, or coach­es knew which con­di­tion the child had been assigned to. Because coach­es were kept blind to chil­dren’s con­di­tion, and thus did not receive detailed infor­ma­tion on how chil­dren were pro­gress­ing through the exer­cis­es, they were unable to pro­vide coach­ing sup­port to par­ents as is done in reg­u­lar clin­i­cal practice.

Out­come mea­sures includ­ed neu­rocog­ni­tive assess­ments, par­ent and teacher reports of ADHD symp­toms, and a glob­al assess­ment of func­tion­ing made by study clin­i­cians. Results indi­cat­ed ben­e­fits of HI train­ing on only 1 of 25 out­come mea­sures, a mea­sure of ver­bal work­ing mem­o­ry. Impor­tant­ly, no train­ing relat­ed dif­fer­ences were found for par­ent, teacher, or clin­i­cian rat­ings. The authors con­clude that their find­ings cast “…doubt on the claims that CWMT is an effec­tive treat­ment in young chil­dren with ADHD.”

Study 2

In a sec­ond ran­dom­ized con­trolled tri­al [Chacko et al., (2013). A ran­dom­ized clin­i­cal tri­al of Cogmed Work­ing Mem­o­ry Train­ing in school-age chil­dren with ADHD: A repli­ca­tion in a diverse sam­ple using a con­trol con­di­tion. The Jour­nal of Child Psy­chol­o­gy & Psy­chi­a­try, 55, 247–253] 85 7- to 11-year old-chil­dren with ADHD were assigned to HI or LI CWMT.

Train­ing con­sist­ed of 5 30–45 min­utes per week for 5 weeks; this is the typ­i­cal ses­sion length for chil­dren in this age range. Addi­tion­al tri­als were added to LI train­ing ses­sions as need­ed so that the length of LI and HI train­ing ses­sions were more com­pa­ra­ble. Also unlike the pri­or study, coach­es had com­plete access to chil­dren’s train­ing data so that they could over­see par­ents as is done in stan­dard clin­i­cal prac­tice using CWMT. Out­come mea­sures includ­ed par­ent and teacher rat­ings of ADHD symp­toms, stan­dard­ized assess­ments of work­ing mem­o­ry, com­put­er­ized assess­ments of atten­tion, and aca­d­e­m­ic achieve­ment testing.

As report­ed pri­or stud­ies, chil­dren receiv­ing active train­ing showed sig­nif­i­cant gains in work­ing mem­o­ry com­pared to con­trol chil­dren. This was true for both visuo-spa­tial and ver­bal work­ing memory.

How­ev­er, com­put­er­ized tests of atten­tion showed no sig­nif­i­cant dif­fer­ence between the groups. The same was true for par­ent and teacher rat­ings of ADHD symp­toms as well as for mea­sures of aca­d­e­m­ic achievement.

Based on these large­ly neg­a­tive results, the authors con­clude that CWMT should not be used as a treat­ment for ADHD.

Sum­ma­ry and Implications

Results from these 2 ran­dom­ized-con­trolled tri­als do not sup­port CWMT as a first-line treat­ment for ADHD. In both stud­ies, there was evi­dence that train­ing pro­duced gains in some non-trained mea­sures of work­ing mem­o­ry. How­ev­er, improve­ments in par­ent or teacher rat­ings of behav­ior were absent. Giv­en the adverse impact of core ADHD symp­toms on aca­d­e­m­ic and behav­ioral func­tion­ing, this is a sig­nif­i­cant limitation.

In the first study, one could argue that coach­es could not use detailed records of chil­dren’s train­ing per­for­mance to guide their coach­ing calls with par­ents, which may have under­mined the train­ing effec­tive­ness. This was not true of the sec­ond study where coach­ing super­vi­sion was pro­vid­ed in the stan­dard man­ner. This sec­ond study was also the largest tri­al of CWMT for ADHD con­duct­ed to date and the sam­ple size was suf­fi­cient to detect mean­ing­ful treat­ment effects if they were there.

What can we con­clude from this work? Despite promis­ing ini­tial reports sug­gest­ing that CWMT is a poten­tial­ly effec­tive treat­ment for ADHD, these stud­ies sig­nif­i­cant­ly under­cut this con­clu­sion. This does not mean that there is no util­i­ty to CWMT, how­ev­er, par­tic­u­lar­ly for indi­vid­u­als with demon­strat­ed work­ing mem­o­ry deficits. If one’s treat­ment goal is to enhance work­ing mem­o­ry, CWMT may have real val­ue. If the goal is to bring ADHD symp­toms under con­trol, how­ev­er, these find­ings indi­cate that for most chil­dren with ADHD, CWMT would not cur­rent­ly be con­sid­ered a rea­son­able sub­sti­tute for med­ica­tion and/or behav­ior therapy.

One final com­ment. I think it is impor­tant to note that many clin­i­cians are using CWMT with chil­dren who have ADHD and many have report­ed that they are obtain­ing good results. A num­ber of these are clin­i­cians that I know and respect, and it is dif­fi­cult to rec­on­cile the neg­a­tive results report­ed here with out­comes that are report­ed by many clin­i­cians who use Cogmed in their prac­tice. This is an exam­ple of where research find­ings dif­fer from clin­i­cal impres­sions, and I don’t think it is pos­si­ble to con­clude with com­plete cer­tain­ty that one is right and the oth­er is wrong. How­ev­er, if one looks to the research to make deci­sions about treat­ments to rec­om­mend for chil­dren with ADHD, rou­tine­ly rec­om­mend­ing Cogmed would be incon­sis­tent with the cur­rent research base in my view.

Rabiner_David– Dr. David Rabin­er is a child clin­i­cal psy­chol­o­gist and Direc­tor of Under­grad­u­ate Stud­ies in the Depart­ment of Psy­chol­ogy and Neu­ro­science at Duke Uni­ver­sity. He pub­lishes Atten­tion Research Update, an online newslet­ter that helps par­ents, pro­fes­sion­als, and edu­ca­tors keep up with the lat­est research on ADHD.

Pre­vi­ous arti­cles by Dr. Rabin­er: