運動障害児に対する神経発達学的治療の効果

-\twza\

eg

22

_tsee

8

e

464

--

471

e

(1995

ff)

Report

TheEffectiveness

of

Neurodevelopmental

Treatment

for

Motor

Impaired

Children*

Meta-Amalysis

for

a

Retrospective

Study

KoichiHIRAOKA**

Abstract

The

purpose of thisstudy was to

determine

the effectiveness of

Neurodevelopmental

Treat-ment

(NDT)

forchildren with motor disabilities.Meta-analysis was used toestirnate the

effec-tiveness through reviewing previous studies.

Meta-analysis

is

a reratively new statistical method that allows us to

find

out overall effectiveness of a therapeutic

intervention

through

reviewing

_previous

research articles thatrelates tothe

issue.

The

effectiveness of the

therapeu-tic

intervention

isrepTesented

by

the effect size thatis

defined

as the

difference

between

the means of the experimental and the control group

divided

by

the contror _group standard

devia-tion.

I

estimated the effect size of the

long

term treatment through experimental studies which

had published between 1985 and 1994

(Study

1).Furthermore, the short term effectiveness of

NDT

was estimated through studies that

had

published

between

1970 and 1994,dealtwith

im-mecliate effectiveness of NDT after one treatment session

(Study

2).The effect size of the Study 1was

-O,

21,and of the

Study

2 was O.69.The results suggested thatNDT intervention was

not _predominant against other control conditions while

the

changes of _performances

during

short term

NDT

interventionwas obvious.

Key words Neurodevelopmental Treatment

(NDT),

Meta-analysis,Developmental

disability

Introduction

Neuro-developmental

Treatment

_(NDT)

has

been

ene of the most _popular ways totreatchildren with

the problems of motor

development,

such as

cere-brai

palsy. However, the effects of the treatment

are controversial.

Many

researchers have failedto

*

_{gyme:wntts#61eeests\eciinuta)imR}

Meta-AnalysiseMVhkNemS}st

" _School

of Physical and OccupationalTherapy,Hakone National Hospital

ytmade

-

:

Nittsasptngeera[nlngpt

_v/N lfvi- y _. y\

M

ve\tstaeFF

(+

2so

reAJI[fiylxWMrtIMza

412)

(9NE

1995

E2

fi

6fi1eeveE 1995

fiE

11

fi

25H)

prove

its

eMcacy

because

of their poor methodologiesi}.

Piper2)

suggested that most of

those studies

had

methodological _problems such as a

lack

of control _groups, sample size, outcome

measures, and sample population.

Because

of these conditions, the effectiveness of NDT for children

with _physical

disabilities

has not

been

established

as yet.

Glass6)

has established a way to

integrate

studies

called "meta-analysis".

This

analysis allows us

to

estimate the raagnitude of effectiveness through

adding up the outcomes of

the

studies. He

Japanese Physical Therapy Association

JapanesePhysicalTherapy Association

The

a strong conclusion on meta-analysis. The analysis

gives us an index which

is

calted effect size,

Effect

size issimply calculated by subtracting means of

control group from means of experimental group

divided

by

standard

deviation

of control group.

There

is

a research _that has used meta-analysis

to estimate effectiveness of

NDT

intervention.

OttenbacherB) attempted to

investigate

effectiveness

of

NDT

through

integrating

studies of

NDT

inter-vention that had been

_published

by

1984

using

meta-analysis. He showed a magnitude of difference

of outcome

between

NDT and other

interventions

in

his

study. This isthe onty study that have

es-timated effect size of

NDT

intervention.

Despite his study, there

is

no research on

es-timating effect size on studies that

investigates

the

immediate effectiveness of an

NDT

intervention.

However, itisa factthat many therapistshave

im-pliedefiicacy of immediate effectiveness of

NDT

in-tervention intheirclinical experience9)iO). DeGaltgii}

suggested the usefulness of the measurement of the short term changes of the_performance to_prove the effectiveness of

NDT.

In

thisreport,

I

investigated

the efficacy of

NDT

through reviewing _previous results of the studies

in

order to determine how much NDT intervention was effective. Meta-analysis was used for

es-timating that.

I

designed

two studies.

One

was the study that

investigated

long

term effectiveness of

NDT,

and the other was the study that addressed

to

find

out short term effectiveness of

NDT.

Method

I

referred

to

the

latestresearch articles that

had

been

published from 1985 to1994 forStudy 1,and

1970

to

1994

for

Study

2.

The

reason I

didn't

search articles published

before

1985 for Study 1

was that a previous study

I

mentioned

before

had existedS) for the articles. The ways to search the

studies were on-line computer searches on Medline,

Cumulative Index toNursing and AlliedHealth

Lit-erature, and manual search on Index Medicus.

Effectivenessof

Neurodevelopmentai

Treatment forMotor Impaired

Children

465

1. Studyl

In

Study

1,

I

gathered articles which met

the

fol-lowing

criteria.

Fjrst,

they

had

to

be

the articles of

the efficacy of

NDT

treatment that

included

NDT

intervention,

Secondly,

they

had

to address the

treatment of children with some kinds of physical

disabilities

such as cerebral _palsy or Down's

Syn-drome. Thirdly, they had to have at least two

groups, a control group

that

was

treated

by

NDT and ano,ther group thatwas not _treated

by

NDT or

by

a reduced

in

intensity

of the

NDT

intervention.

Fourthly,

they

had

tomeasure the effectiveness of

the accumulation of more than one therapeutic

in-tervention. Finally, they had to have dependent

variables of at ]eastmean and standard deviation,

or F,or tso thatIcould calculate "effect

size".

2.

Study2

In

Study

2,

I

gathered articles that met the

fol-lowing criteria.

First,

they

had

tobe the articles of

the efficacy of

NDT

treatment that

included

NDT

intervention. Secondly, they might address the

treatment of children with some _physical

dis-abilities such as cerebral _palsy or

down

syndrome.

Thirdly,

the articles had tocornpare the

functional

status of pre-treatrnentand post-treatment

measur-ing

the

immediate

carry-over effects of single NDT session. Finally,they

had

tohave dependent

varia-bles

of at leastrnean and standard

deviation

or F

or tvalues so thatIcan calculate `teffect

size".

3,

Hypothesis

I

established a

hypothesis

for

Study

1

that

NDT

intervention

had

a

large

advantage over the other

eonditions. Also I established a hypothesis that

there was a short

term

effectiveness of

NDT

inter-vention for Study 2. For these hypetheses, I

as-sumed

that

both estimated effect sizes should

be

large.

4. Analysis

AsI

mentioned

before,

I

used rneta-analysis te

466

eeIittstaY

tiveness of NDT.

Control

grQups' scores were used

as the control conditions of

Study

1.

Pretest

scores were used as scores of a control condition inStudy

2

because

they

didn't

have

a control condition.

Al-though the way in Study 2 isnot the formaf way

for meta-analysis, it is_permissible to use

meta-analysis for single or multiple subject designs that don't have control _groups. Because Glassii)said

that

experiments

that

comprjsed only pretesting

and pesttesting

in

a single group of subjects might

be included

if

the treated group's members' pretest

status was a good estimate of theirhypothetical

post treatment status

in

the absence of treatmenL

Because

of thiscondition, effect size of

Study

2 is supposed toshow the differencein_performance

be-tween pre-test and post-test of each subject within a short term.

In order toavoid biased estimate of effect size,

effect size was multiplied

by

corrected

factor,

Also

inorder toavoid

bias

of homogeneity, _pooled

esti-mate of variance was used

for

_population variance.

Weighted

mean was estimated so as toestimate the overall mean effect sizei2).

I

summarized an

exam-ple of the procedure estimating the effect size on

Appendix.

Results

Between 1985 and 1994,

there

were 11 studies which had investigated effects of NDT on children

with physical

disabilities.

Out

of the 11 studies,

there were

6

experimental designed studies. Two

studies were excluded

from

the studies

because

of

lack

of control group, and a study was excluded

be-cause of

lack

of description of standard deviation.

Therefore,therewere

3

studies thatmet thecriteria

I had established forStudy 1.

Between

1970

ancl

1994,

there were 20 studies which

had

investigated

effects of

NDT

for

children

with physical disabilities.

Out

of the 20 studies,

there were

7

studies investigating the short _term

effectiveness of NDT. One study was excluded from

the studies

because

of

lack

of control group, and

three studies were excluded

because

of lack of

de-eg

22

_geg

8

_e

scription of standard

deviation.

Therefore,

there

were

3

studies that met the criteria

I

had

tabilishedfor

Study

2.

Studies

I

referred were

listed

on

Table

1.

Ilistedindividualeffect size on Table 2.0verall

mean effect size in Study 1

ivas

-O.21.

0verall

mean effect size in

Study

2

was O,69,The summary

of information of thestudies were listedon Tables

2

and

3.

Discussion

As

I mentioned above, the result

in

_{Study 1} showed

-O.21

of effect size.

The

effect size is

sidered as small effect size according to

Cohents

criteria27). Cohen suggested that

.2

was considered

as small effect size,

.5

was considered as medium

effect size which was largeenough tobe visible to

thenaked eye, and

.8

was the largeeffect size.

Accordingly,

Iconcluded

that

NDT

approach was

not _predominant or slightly

detrimental

against

other conditions. Therefore, the

hypothesis

for

Study 1 was rejected. In

Ottenbacher's

study,

he

showed effect size of

NDT

intervention

as

O.31,

which was

interpreted

as children who had

ceived NDT _performed slighLly betterthan children who hadn'treceived NDT.

The difference between the results of

Ottenbacher's

study and

the

study presented here

may

be

caused by differenttrends of the outcome

of the studies, Thirty-sixpercent of thestudies

sented after 1984 showed _positive effectiveness of

NDT

while 56 percent of the studies presented

tween

1970

and

1984

showed positive effectiveness

according tothe 20 articles referred to inStudy 2.

That

implies

more negative trend

is

_predominant

in

the

later

studies.

As

theresult of

Study

2,

I

had

O.69 of effect size

which was considered as a

large

effect size

ing

to

Cohen's

criteria.

The

effect size suggested

that NDT intervention causes children to _perform

much better than they

_did

on _{pre-intervention}

within a short period,Therefore,the

hypothesis

for

Japanese Physical Therapy Association

JapanesePhysicalTherapyAssociation

The Effectivenessof

Neurodevelopmental

Treatment forMotor Impaired

Children

Table 1. Research articles

that

relate

to

NDT outcorne foryouth population

467

ResearchersStudy#Year Subjects ResearchDesign MeasurementTools Wrighti3) 1973cerebralpalsy Experiment originalscores

Carlseni4)

1975cerebralpalsy

Experiment

BayleyScate,DDST

Norton15) 1975multiple Singlesubjectfrequencyofbehavior

Rothmani6)

41978cerebralpalsy

SinglesubjectVC,PVC,PPVC,FEV

Avignoni7)

1981riskinfants

Experiment

physicalexamination

Harris9)

1981Down'sSyndromeExperiment

BayieyScaLe,PeabodyScale

Sommerfeld5)

1981cerebralpalsy

Experiment

Wilsondeyelopmental

RefiexTest

WCISDGrossMotorEvaluation

ROM

Jenkinsia)

1982developmental disability Experirnent Peabodyscale

DeGangii)

1983cerebralpalsy Experiment

ADLtest

Goodman19) 31985riskinfant Experiment

GriMth'smentaldevelopment

scale

Piper2)

1986riskinfant

Experiment

WoianskiGrossMotorEvalua-tionWilsonDevelopmentalReflex

Profile

Milaniscreeningtest

Griffth'sMentalDevelopment

Scale

Herdon20) 1987cerebralpalsy

_{QuasiExperiment}

movementpattern

Palmer2i) 11988cerebralpalsy Experiment

BayleyScale

Piper2Z)

1988riskinfants

QuasiExperiment

neurologicalsymptoms

Edwards23)

1989Down'sSyndromeSinglesubjectBayleyScale

Lillyle)

1989cerebralpalsy

SinglesubjectADLtest

Srnelt24)

1989cerebralpalsy

SinglesubjectROMhandsurfacearea

Kluzik25)

5'I990cerebralpaLsy

SinglesubjectMovementTime(MT)

MovementUnits(MUs)

PercentageofMTofFirstMUs

Mayo4)

21991cerebralpalsy

Experiment

BayleyScale

Chakerian26)

61993cerebralpalsy

Singlesubjecthandtracingarea

'The

study numbers correspond tothe study numbers on Tables2and 3.

suggested that

NDT

intervention

made a

large

change

irr

_perforrnance within a short terrn

al-though the

NDT

intervention

might

not

be

significantly

different

from

other conditions

in

terms of

long

term

effectiveness.

I

suppose that the reasonsI have had Iargeeffect size

in

Study

2

are the

length

of the period of the

468

Table 2.

ve\tante}g

eg

22

_tseg

s

_e

Summary

Informatien

for

the

Study

1

Study# rnean(E)

SD(E)

N(E) mean(C)

SD(C)N

(C)ESWeighted

Corrected Corrected

Sp Factor ES 1111112333333

49.1

65.6

67,8

47.9

67

67.5173.4

98

102

101 101

96

95

11.616.3

1918,918.423.299.4

11

20 10 12 13 13

25252525252517202020202020

58.I75.572.563.375,376.6159.5

98

102

98

96

95

97 1914.918.2

2215.318.382.3

13

18 11 16

13

18

23232323232312202020202020

･-o.-o.-o.-o.-o.-o.

o.

o,

o,

o.-o.

58632575494315oe29350813

15.5815.6518,6220,4416.99

2192.8112.0419,0210.5114.14

13 15.7

O.98O

98O.98O,98O,98O.98O.97O.98O.98O.98O.98O.98O.98

-o.-o.-o.-o.-o.-o.

o.

o, o.

o.-o.

57622574484351oo28350812

ES=effectsize.

(E)=Experimentgroup,

(C}=,-Controlgreup.N=Number ei samples,

Table 3.Summary IRformation

for

the

Study

2

Study#

rnean(E) SD(E) N(E) mean(C) SD(C)N

(C)ESWeighted

Corrected

Corrected

Sp

Factor

ES

44445556 1.32 O.13 1.36

O.13

O.94

O.05

O.94

O.03

1267,25

318.11

2.73 1.48 37.58 16.56 16.75

3.33

555545519 O.86

O.13

1.36

O.13

O.63

O.05

O.94

O,03

1691.75 566,09 3.22 1.23

26.2

15,76

14.58

3.21

5555455193.48

o6.46

oO.92O.36

O,7O.66 O,13

O,13

O,05

O.03459.16

1,36

16.17

3.27

O.9

O,9

O.9

O.9O,87 O.9

O.9O.98

3,14

o5,84

o O.8O 33O.64O,65

ES==effectsize.

(E)=Experiment

group.

(C)=Control

group. "ES

values in ES#5 and 6 are po$itivealthough means of

theexperimental _groups subtracted by means of the centrol groups are negative. Becausethe negative values mean

posi-tiveresults inthose values.

can see on

Table

1,

100%

of studies that measure short Lerrn effectiveness ef NDT used evaluative

measurements made

for

rneasuring the outcome of

treatments specifically. They measured

qualitative

changes of movements

based

on therapy's target specificity. Some therapists9)iO)have

proposed

utiliz-ing of measurement toolswhich are

based

on

indi-vidual treatment _goals.Parette28)said that a more

comprehensive assessment batteryadministered on

a _{pre-treatment/post-treatment}

basis

may more ac-curately reflect the influence of therapeutic

inter-vention. So,although there

is

no standardized

test-ing to evaluate short term effectiveness of

treat-ment

for

children with cerebral palsy, measuring

qualitativechanges of theirmovements must

be

a $ignificant way toevaluating the outcornes of

ther-apeutlclnterventlon.

On

the other

hand,

other studies mainly used

di-agnostic measurernent tools,such as

Bayley

Scale

and

Peabody

scale, that may not

be

sensitiye

enough todetectthe outcomes of interventions.In

Study

1,67% of thestudies used

Bayley

Scale

for

the assessment.

This

is

rnacle

for

discriminative

measure, and no responsiveness

data

has

been

pre-sented to establish the Bayley

Motor'

Scale

as an evaluative instrument29)30).stern3) said that the

Japanese Physical Therapy Association

JapanesePhysicalTherapyAssociation

The Effectiyenessof

Neurodevelopmental

Treatment

for

Motor Impaired

Children

Table

4.The

Relationship

between

Opinions

on thestudy outcomes

and the Length of the Intervention

469

<1 month1 _<3 monthS-months3 monthsS <6 months6 monthsS PositiveEffects

No

changes

41

32

13 13 'The

numerals intheblanksindicatethenumber ef research articles corresponding to thecolumns and thelines.

cause ef theobscure effectiveness of NDT

interven-tion is that rnost standardized scales, such as

Bayley

Scale,

didn'taccount forthe quality of the

movement although measuring the qnality of

movement was the most

important

framework

in

determining

the

improvement

ot the motor per-formance. Lillyie)also said that the measurement

criteria used

in

those studies were not sensitive

enough to detect small changes occurring

due

to

NDT,

Also,

length

of _period of intervention isan

impor-tant

factor.

As you can see in Table 4, there isa

trend that the results of

the

reports worsen as

the

length of _period of treatment increases.Based on

that trend,Imay be able to

interpret

the result of

Study 2 that I can see obvious predominancy of

NDT as lengthof _periodof treatment

is

short.

The limitationof thisstudy was small sample

size. Glassii}said that more studies are

better

than

fewer.

So,

the

number of studiesIreferred tofor

es-tirnationof effect size rnay weaken the conclusion

ofthisstudy.

APPENDIX

Showing an example of estimating effect size as follows,

The values described belowwas sited from the firstlineof

Table2.1.

Calculatingthe square root ef the weighted _pooled

varlance i

(NE-1)SZ+(Nc-1)SZ

Weighted SP=V N,+Ncu2

v(25-1)IL62+(23-1)19.o2

25+23-2

-15.58

2. Calculating _{theCorrectedFactor}

3 CorrectedFactor=.'1-

--

4m-1 3 4(25+23-2)--1 =.O.98 'm=NE+Ncm2 3. CalculatingCorrectedEffectSize

Es=

g-:mC.

=

49.1-58.1

=-o.ss

Sp 15.6

CorrectedES

==

ES'Corrected Factor=:

-O.58'O.98

==

-O.57

References

1)DeGangi GA, Hurley L,et al,: Toward a methodology of the short-term effects of neurodevelopmental

rnent. Am _JOccupTher ₃₇

_(7):

479-484. _1983,

2)PiperMC, Kunos I,et aL: Early physical therapy

effects on the high-risk infact)A randomized trolledtriaLPediatrics78 (2):216-224, 1986.

3) Stern FM, _{Gorga D: Neurodevelopmentaltreatment}

(NDT);

Therapeuticinterventionand itseMcacy.

fantsand Young Children1

(1):

22-32,1988.

4)Mayo NE: _The effect of _physicaltherapy for childre]

with motor delay and cerebral palsy; A randomized

clinical trial.Am jPhys Med Rehab 70

_(5):

258-267,

1991.

5)Sommerfeld _D,FraserBA,etaL: Evaluationofphysical

therapy service for severely mentally impaired dents with cerebral palsy.Phys Ther 61

(3):

338-344,

1981.

6)GlassGV: Meta-AnalysisinSocialResearch. Sage

lications,1981.

7)Thomas JR: RcsearchMethods in Physica} Activity.

Human KineticBeoks, 1990.

8)OttenbacherKJ,BioccaZ,et at.:Quantitativeanalysis

of the effectiveness of _pediatrlctherapy; Emphasis en

the neurodevelopmental treatment approach. Phys

Ther 66 (7]:1095-1101,1986.

9)Harris RS: Effects of neurodeveloprnental therapy on motor performance of infantswith Down's syndrome.

Dev Med Child Neurol 23:477-q83, 1961.

10) _{LMy LA, Powell NJ: Measuring the} effects of

neurodevelopmental treatment on the daily living

skills of 2 children with cerebral palsy. Am

J

Occup

Ther 44

<2):139-145,

1990.

470

ve#tsza\

research. Review of Research in Education 5:351-379, 1977.

12) Themas JR,French KE: The use of meta-analysis in

exercise and sport] A tutorial.Res

Quart

ExerSport57

(3):196-204, 1986.

13) Wright T, Nicholson _J:Physiotherapy forthe spastic

ehild: an evaluation. Dev Med Child Neurol 15:

163,1973.

14) CarlsenPN: Comparison of two occupational therapy appreaches for treating the yeung cerebral palsied

child, Am JOccup Ther 29

{5):

267-272, 1975.

15)NortenY:Neurodevelopment and sensory integratien. Am JOccup Ther29

(2):

93-100,1975.

16)Rothman

JG:

Effectsof respiratory exercises on the

vital capacity and forcedexpiratory volume in

dren with cerebral _palsy,Phys Ther 58

{4):

421-525,

1978.

17)Avignon M, Noren L,et al.: Earlyphysiotherapy Ad

Medum Veita or Bobath in infantswith suspected

neuromotor disturbance. Neuropediatr 12(3):232-241,

1981.

18)_jenkinsRj:Effectsofdevelopmentaltherapy on rnotor impairedchildren. Phys Occup TherPediatr,2,4,1982, pp19-28.

19)_Goodman M, Rothberg AD: Effect of early neurodevelopmental therapy in normal and aL risk survivors of neenatal intensivecare. Lancet14: 1330, 1985.

20)Herdon WA, Troup P,ei al.:Effects of mental treatrnent on movement patterns Df children with cerebral palsy.

J

PediatrOrthe 7

_(4):

395-400,

1987.

21)Palmer FB,ShapjroBK,et al.: The effect$ of physical

therapy on cerebral _palsy;A controlled trialin infants

ag

22

igee

8

-g

with spastic dip]egia.New Eng JMed 318

(13):

808,1988.

22> Piper MC, Mazer B,et al.: Resolution of neurological symptoms in high-riskinfantsduring the firsttwo

years oflife.Dev Med _ChildNeurol _30:26-35, _1988. 23) Edwards SJ,Yuen HK: An intervention program fora fraternaltwin with Down syndreme. Am

J

Occup Ther 44

(5):

454--458,1990,

24) Srne]tHR: Effect of an inhibitiveweight-bearing mitt on tene reducation and functionalperformance ina child with cerebral palsy.Phys Occup Ther Pediatr9

(2):

53--80,1989.

25)Kluzik J:Quantificationof control; A preliminary

study of effects on neurodevelopmental treatment on reaching inchildren with spastic cerebral palsy.Phys

Ther 70

(2):

65--78,1990.

26)ChakerianDL, LarsonMA: Effectsof upper-extremity

weight-bearing on hand-opening and prehension

terns inchildren with cerobral palsY.Dev Med Child

Neurol 35:216-229, 1993.

27)Cohen J:StatisticalPower AnalysisfortheBehavioral

Services.

Academic PressInc.1977.

28)ParetteHP,Hourcade

JJ:

A review of therapeutic

vention research on gross and finemotor progress in

young children withcerebral palsy,Am

J

Occup Ther

38 (7):462-468, 1984.

29)Deyo RA: Teward cEinical apprications ef healthstatus measures: Sensitivity of scales toclinically important changes. Health ServiceResearch 19

{3)/

275-289,

1984.

30) Rosenbaum PL, Russell DJ : Issues in measuring change in motor function in children with cerebral

palsy:A special communication. Phys Ther 70 (2)i125

Japanese Physical Therapy Association

Japanese _{Physioal 　Therapy 　Assooiation}

The

　

Effectiveness

　of　

Neurodevelopmental

　

Treatment

　

for

　

Motor

　

Impaired

　

Children

471

〈要 　 　 約 〉 運動障害児に対する神経 発 達 学的治療の効 果

一

_Meta

_−Analysis

_{を用いた文 献 分 析}

一

平 　岡　浩

一

国立療 養所箱根 病院附属リハ ビ リテ

ー

シ ョ ン学 院　理学 療法学 科 （〒

250

神 奈川県小田原市風 祭

412

） 　こ の研 究で は

，

運 動 障 害 児に対する神 経 発 達 学 的 治 療 （NDT ）の効 果につ い て検 討した

。

過 去に発 表された神 経 発 達 学 的 治 療の効 果に関 する文 献 を もとに Meta

−Analysis

を用い て全 体として の治 療 効 果の _大き さ を判定し た。 Meta

−

Analysis は過去の研究論 文を 通し てある治療 介 人の効果を測定 する比 較的新しい_{統 計学}的方法で ある

。

その治 療 介 入の効 果は

，

実 験群の _平均値か らコ ン トn

一

ル_群の _平均値 を減算し

，

その_値をコ ン トロ

ー

ル _群の_{標 準偏}差で_除算し た数 値 （

Effect

　

Size

_）に よっ て表さ れ る。 　 1985年か ら1994年の_間に発 表さ れ た 研究 論 文を も とに

NDT

の _{長 期 的 治 療 効 果 〔複 数 回 治 療}の_{累 積} 効 果 ）にっ い て の

Effect

　

Size

_を算 出_{し た} （研 究

1

） 。 更に

NDT

の短期的 治 療 効 果 （

1

回 治療の直 接 効 果） にっい て 1970年か ら1994年の間に発 表さ れ た研 究 論 文を も とに Effect　 Size_{を算 出し た　（研 究} 2）

。

　 研究

1

お よび研 究

2

の_{文献 採}択 基準に合 致 する論 文 数は それぞれ 3であっ た_。 長期 的 治療 効 果の

Effect

　

Size

（研究 1）は

一

〇

．

21

，

短 期 的治 療 効 果の Effect　Size （研 究2）はO

．

69であっ た_Q 本研究の

結果よ り， 運動障害児に対する神経発 達学的治療は

，

短 期 的 効 果は認め ら れ る が， 長期 的に は他の治療

介 入と 比較して ほ ぼ同 等

，

あ るい は わ ず か に そ れ ら よ り低い治 療 効 果を 示 す事が示唆 さ れ た。