Neurofeedback for Insomnia

Report
A New Look at an Old Workhorse:
Sensorimotor Neurofeedback
Barbara U. Hammer, Ph.D., Agatha P. Colbert, M.D., Kimberly A. Brown,
MSOM, Helfgott Research Institute, National College of Natural
Medicine, Portland, OR, Elena C. Ilioi, Psychology Honours, McGill
University, Montreal, Quebec, Canada,
The authors wish to thank the Helfgott Research Institute of the
National college of Natural Medicine for its support of this
research, and William L. Gregory, Ph.D. for providing our
statistical analysis. We appreciate of the generous time, energy,
and support provided by our Research Assistants who helped in
the development and conduct of the investigation: Sean E.
Griffith of the Psychology Department at Duke University and
Tineke Malus of the Natural College of Natural Medicine. This
study was not supported by any industry funding.
2005 NIH Conference on Insomnia declared Insomnia
an Epidemic:
 10-20% of adults, mostly Women & Seniors, have
Insomnia
 Another 20-40% report sleep disturbances
 40% of children, especially those with other
vulnerabilities
National Sleep Foundation--November, 2009
•27% say their sleep has been disturbed at least a few nights a week in the
past month due to personal financial concerns (16%), the U.S. economy
(15%), and/or employment concerns (10%).
•54% of those losing sleep over economic concerns also had difficulty with
their feelings at least a few days a week in the last month.
•People with sleep problems are sleeping less than 6 hours on a typical
workday or weekday (35% vs. 14%); and/or
•They say they’ve driven drowsy at least once a
year (41% vs.
23%).
month in the past
Problems resulting from Insomnia are increasing
National Sleep Foundation 2009 poll:
Effects on Health and Daytime Funcitoning:
Compared to their better sleeping counterparts, these people are
more likely to report being unable to do the following
because they are too sleepy:
o Work well and efficiently (25%)
o Exercise (30%)
o Eat healthy (22% )
o Have sex (16% )
o Engage in leisure activities (28% )
more Effects on Health:
• Memory formation impairment=long-term
memory impairment (Buzsaki, Sept.,2009)
•Additional health care costs up to $14 billion (NSF)
•Sleepy drivers =major public safety problem &
cost(NSF)
•Higher work place accident rates & lost productivity
estimated $80 billion (NSF)
•Reduced immunity if Sleep Efficiency <92%
(Cohen, Oct., 2009)
October, 2009 (Cohen, 2009) found immunity
to rhinovirus correlated with the amount of
sleep in healthy adults.
1. Graded association with average sleep
duration: participants with <7 hours of sleep were
2.94 times (95% confidence interval [CI], 1.18-7.30) more
likely to develop a cold than those with 8 hours or more
of sleep.
2. The association with SE was also
graded: participants with less than 92% efficiency were 5.50
times (95% CI, 2.08-14.48) more likely to develop a cold
than those with 98% or more efficiency
Insomnia Definition
Primary Insomnia (DSM 307.44): Complaints of Difficulty Falling
Asleep, Staying Asleep or Awakening too early, or Non-restorative
Sleep which occurs for at least one month duration and:
1. Causes significant distress or impairment in social,
occupational, or other important areas of daytime functioning.
2. Does not occur exclusively during the course of
Narcolepsy, Breathing-Related Disorder, Circadian Rhythm Sleep
Disorder or a Parasomnia.
3. Does not occur exclusively during the course of another
mental disorder.
4. Is not due to the direct physiological effects of a
substance or general medical condition.
Definition of Insomnia Most Widely Used
In Clinical Practice (American Board of Family Practice) and
often in Research:
1) Sleep Latency (SOL) >30 minutes
2) Sleep Efficiency (SE) <85%
3) Sleep Disturbance >3 times a week
4) Research (Sleep, 2006, April: 29(4)) suggests cutoff of 20” is
most sensitive and specific.
TREATMENTS:
Pharmaceuticals
•Benzodiazepines most widely used, cannot use long term
•Hypnotics can also cause Depression
•Reports often hide drug harm (Lunesta, Rozerem)
CAM:
•In 2007, 4.5% (1.6 million) of those surveyed used CAM
treatments for Insomnia. Summary from NCCAM
website November, 2009:
CAM
•Herbal products studied lack demonstrated
evidence of effectiveness:
•Valerian, Melatonin, Chamomile, Kava, 5HTP, & L-tryptophan (may be harmful, banned)
•Aromatherapy-some sleep inducing effects
Other
CAM
•Acupuncture-inconclusive.
New RCT shows promising results,
from 270” electroacupuncture. Only slight but sign. difference pre vs
post Rx in SE from Diaries and from Actigraph.
•Meditation, yoga-inconclusive. Just completed, unpublished RCT :
daily Kundalini, just prior to bedtime, 8 wks>sign. reductions in severity.
•Relaxation techniques—EMG Biofeedback etc=
mixed/inconclusive. Unpublished study showed SMR >EMG
Psychological treatments
•highest co-morbidity
• Insomnia persists despite psychotherapy for
depression or anxiety
•Physiological hyperarousal hypothesis-some support
• Cognitive Behavior Therapy—Demonstrated Efficacy
•Sleep Restriction
•Stimulus Control
•difficult to administer
•New Internet based program promising
1976, Hauri et al:
Waking SMR correlated +.64 with SE &
-.64 with SOL
1981, Hauri et al:
Compared EMG & Theta & SMR biofeedback
Only SMR significantly improved Sleep
Tense Insomniacs benefited from Theta BFD
September, 2008 aapb:
Peter Hauri (Mayo Clinic): SMR Neurofeedback in
1980’s used Analog Equipment not feasible
for general clinical use:
Too Expensive
Too Cumbersome
Too Time consuming
“Time to revisit SMR for Insomnia with Digital
equipment and new training methods.”
This was the exact purpose of our study, which
had been given IRB approval just the previous
month-in August, 2008!
Overview:
Purpose –Compare effectiveness of SMR NFB and (a
sequential, quantitative EEG) an EEG guided
(IND) protocol for amelioration of Insomnia
Methods –RCT single-blind study.
Intervention –Groups received 15 20-minute sessions
of Z-Score NFB.
Pre-post measures –Mental health (MMPI-2-RF and PDSQ),
Quality of Life Index (QOLI), Insomnia Severity
Index (ISI), Pittsburgh Sleep Quality Index (PSQI),
sQEEG.
Neurofeedback Background & Rationale:
2 Primary Approaches Reflected in our 2 Groups
1. Symptom Based:
Early SMR
Othmers’ 1980’s
2. 2-19 Channel QEEG/Brain Map Guided
Lubar
Walker
Ayers
PARTICIPANTS:
20+ Telephone Screening –unpaid, recruited over 4 months
Exclusions—use of sleep aids, psychotropic meds, mental or
physical disorders that could interfere, prior NFB,
enrolled in another sleep study, pregnant
15 Passed Telephone Screening
12 Invited to enter study
2 Declined to start- personal/extraneous reasons
2 Dropped out- personal/external reasons ≤8 visits
Figure 1: Demographics, Complaints, & Scores
Age Sex
61
58
50
54
50
49
34
40
F
F
F
F
F
M
M
M
Problems
PreISI PostISI PrePSQI PosPSQI
WASO,WE
18
SOL,WASO,WE 28
WE,SOL
12
WE,WASO,SOL 14
WASO
15
NS,WASO,WE
14
SOL,WASO,WE 19
SOL,WASO
17
5
7
7.5
6
9
9
8
1
Insomnia/Sleep Disturbance Cutoffs =8= ISI
14
17
11
9
11
12
17
16
3
5
5
4
5
5
6
3
Duration
Childhood
Childhood
22+ yrs
20+ yrs
1-5 yrs
10+ yrs
5 yrs
1 yr
5=PSQI
Pre Average: ISI= 17.13 PSQI= 13.38
Post Average: ISI= 6.56 PSQI= 4.5
Group
SMR
SMR
IND
IND
SMR
IND
SMR
SMR
MEASURES continued
MMPI-2 RF--Most widely researched, used measure of
psychopathology. Newest version
Psychiatric Diagnostic Screening QuestionnairePDSQ—Guide to depth clinical interview to
confirm absent Dx
Quality of Life Index-QOLI—Measures positive mental
health=daytime function
Actiwatch--Records movement/lack of movement. 3
days pre vs post. Multiple technical difficulties
prevent use of data.
MEASURES continued
sQEEG—sequential EEG—EEG Screening
•
Records several minutes at 4 scalp sites in
succession in 5 runs
•
Records connectivity measures between
each set of 4 sites
•
Certified Calibration tested EEG amplifier for
NFB/sQEEG for FDA
METHODS:
NFB=Operant Conditioning to Norm
•Training to Norm =Normalizing physiological
process via self-regulating brainwave
distribution
•Based on Principles of Learning via Reinforcement
•Need to monitor progress continually
•Z Score NFB designed to use Live, Instantaneous
record as basis of Reinforcement
Z Score Neurofeedback
Physiological training toward Norm Z=0
Results
Insomnia Severity Index
Mean
ISI-PR Total
ISI-PO Total
Std. Deviation
N
16.19
5.345
8
6.56
2.638
8
•PR = pre, PO = post
•Age and Gender were not significant and did
not interact with anything so dropped
•Pre to Post Change Test F(1,6) = 18.2, p < .005
Group
ISI-PR Total IND
SMR
Total
ISI-PO Total IND
SMR
Total
Mean
Std.
Deviation
N
13.17
1.041
3
18.00
6.205
5
16.19
5.345
8
7.50
1.500
3
6.00
3.162
5
6.56
2.638
8
There is a hint of an interaction, p < .18, change differs between
the treatment groups. The IND group does not change as much
as the SMR group. It suggests IND is certainly not better than
SMR, and it may be the other way around.
PSQI
Mean
Std.
Deviation
N
PSQI-PR
Total
13.75
3.495
8
PSQI-PO
Total
4.50
1.225
8
Pre to Post treatment Time change is sig, F(1,6)
= 55.6, p < .0001.Covariates not significant
PSQI
Group
PSQI-PR IND
Total
SMR
Total
PSQI-PO IND
Total
SMR
Total
Mean
10.67
15.60
13.75
4.50
4.50
4.50
Std.
Deviation
1.528
2.966
3.495
.500
1.581
1.225
N
Time by group interaction is marginally significant,
F(1,6) = 4.5, p < .08. SMR is better.
3
5
8
3
5
8
PSQI Sleep Efficiency
Mean
Std.
Deviation
N
PSQI-PR
SE
77.638
11.1530
8
PSQI-PO
SMR
93.18
5.251
8
Pre to Post Change over time is sig, F(1,6) =
15.8, p < .007. Covariates not significant
PSQI Sleep Efficiency
Group
PSQI-PR IND
SE
SMR
Total
PSQI-PO IND
SE
SMR
Total
Mean
Std.
Deviation
N
81.533
1.6623
3
75.300
14.0743
5
77.638
11.1530
8
93.80
6.521
3
92.80
5.149
5
93.18
5.251
8
No interaction with treatment. No treatment is
better than the other.
QOLI:
QOLI-Pre Total
QOLI-PO Total
Std.
Mean Deviation
46.13
12.889
52.63
11.211
N
8
8
Pre to Post Time change is significant, F(1,6) = 9.6,
p < .02. Covariates not significant
QOLI-PR
Total
Group
IND
Mean
Std.
Deviation
N
53.33
10.970
3
41.80
12.969
5
46.13
12.889
8
56.67
7.506
3
SMR
50.20
13.122
5
Total
52.63
11.211
8
SMR
Total
QOLI-PO
Total
IND
Pre to Post Time by group interaction tends toward sig, p < .23.
SMR slightly better.
QEEG
Abnormal
Waves
Delta
Beta
Hi Beta
Pre
107
54
21
Post
42
33
17
Significance
p<.001
p<.01
p<.11
Delta: Pre to post changes 107/304 vs 42/304 yields Z =6.0, p < .001, signif.
Beta: Pre to post 54/304 vs 33/304 yields Z = 2.6, p < .01, significant
Hi Beta: Pre to post not significant, p < .11, not significant
MMPI-2-RF: Pre
vs Post
Changes in Borderline Profile -004
Overall T score: Pre=54.11, Post=50.56. Three excessive Scale scores
reduced to Normal: Demoralization, Dysfunctional Negative, Aberrant Ex
MMPI-2-RF: Pre
vs Post
Changes in Borderline Profile -010
Overall level Pre=73.3, Post=50.22 yielding >23 pt.
reduction=very significant statistically &
clinically
Actiwatch
Multiple Technical Difficulties Prevent Analysis
•Click sound inaudible
•Possibly Defective recording hardware
and software
•Corrections of data with Sleep Log questionable
•Participant errors
•Processor errors
Summary of Results:
•Total Scale scores Significant on both Insomnia Measures: ISI=p<.005,
PSQI=p<.0001
• PSQI Sleep Efficiency improved significantly p<.007
•QOLI improved significantly <.02
•Slight but not significant tendency for SMR to be better than IND.
• Abnormally high levels of Delta and Beta power at Baseline p<.001
•Delta and Beta power were significantly lowered post-treatment p<.001
Baseline HiBeta amplitude was not excessive.
•MMPI-2-RF, borderline-normal subjects showed post-treatment clinical
improvement.
Conclusions:
1. Baseline EEGs showed both excessive sleepiness and
hyperarousal, which significantly improved posttreatment.
2. Both NFB protocols provided significant
improvement in self reported sleep and daytime
functioning.
3. Some tendency for SMR treatment to be more
effective than IND, and it was significantly less
burdensome to administer.
4. No adverse events reported from full treatment
Discussion
1) Data replicates early SMR studies with new
equipment and advanced software/training
designs
2) Z score NFB possibly effective after <15 Rx
sessions (≤300”)
3) SMR may be more effective than Individually
designed protocol based on sQEEG
4) All participants improved on ALL self-report sleep
measures
5) The power of Z Score SMR NFB suggest it could be
very useful for the general Insomnia population
Discussion (continued)
5) sQEEG improvement demonstrates daytime
physiological normalization
6) All subjects tolerated 9 weeks of Z Score NFB
7) 15 sessions of NFB safe
8) Non-Invasive, non-pharmacological
9) SMR Easily replicated and practical for clinic use
10) Robust effects
11) QEEG or sQEEG may offer useful Biomarkers for
Insomnia

similar documents