Further to my review of The Lancet Neurology report . . . :

The report ". . . patient population in the present study is characteristic of the ALS population in terms of age, sex, site of onset, and proportion with a clear family history . . ." is erroneous insofar as gender is concerned . . . no publication I am aware of has males nearly four times more likely to acquire ALS than females. The idea of randomization in the context of these kind of trials is to produce what could be called 'fair teams' in the groups undergoing treatment in consideration of whether there is efficacy against the disease(s) processes . . . one of the problems of ALS research is the unpredictability of each particular patient's likely survival. Would equal unfairness improve the reliability of a result? Variances as high as those found in ALS are quite rare among diseases, if there are any existing at all. The easiest tactic to compensate for this problem is to make the 'teams' larger, probably much larger; in a group of thirty-seven patients one would expect about seven whose regression rate is slow. "The Lancet Neurology's" report indicates five plus one patients had effectively abandoned the DPS branch of the trial. Could all of these have been slow regressors? How vulnerable are these trials to disruption . . . "to death from any cause."

Why did they call it ALS rather than MND (in the UK they call it MND)?

I have no print copy of the report . . . although I am searching for one. To the best of my knowledge and belief the French have not yet published a full report.

These URLs the assessments included in the DiPALS report.

SAQLI (Sleep Apnoea Quality of Life)
http://www.thoracic.org/members/asse...ires/saqli.php

SF-36 (Short Form Health Survey)
http://www.sf-36.org/tools/sf36.shtml

EQ-5D
http://www.euroqol.org/about-eq-5d.html

EQ-5D-3L (Euroqol 5D questionaire 3-level format)
http://www.euroqol.org/eq-5d-products/eq-5d-3l.html
Patient
Health State
Thermometer scale
Carer
Health State
Thermometer scale

I thought I should suggest listening to the audio commentary associated with the DiPALS report . . . you might have overlooked the audio recording of Chris McDermott and Richard Lane discussing the DiPALS study. The interesting audio commentary is found adjacent to the report on The Lancet Neurology's web site and adds content not found in the report:

http://www.thelancet.com/journals/la...152-0/abstract

UK's MNDA comments on DiPALS results:

https://mndresearch.wordpress.com/20...mnd/#more-3701

I am nearing the completion of my review of the DiPALS Writing Committee report . . . sorry for the delay.

I would like to encourage SiTraN's DiPALS Writing Group to conduct a study on the " . . . web-based randomisation system. . . " using the remainder of their applicants . . . their report indicated there were 759 applicants, whereas they only needed 108 for their study. Such would result in about a dozen or so groups, wherefrom one can derive survival statistics . . . then consider whether the several groups developed similar or dissimilar survival profiles. Generally, PREDICTING the survival of patients is considered difficult to do, some researchers even say impossible and such is a primary issue in the development of sound treatment trials. Will the ". . . web-based randomisation system . . ." produce groups with similar or dissimilar survival profiles? Such study may produce a basis for a journal paper all by itself.

Although I continue reviewing the DiPALS report from SiTraN, I have another observation to note:

Again reviewing the numbers along the lower horizontal edge of the graph:

37 36 26 13 6 4
37 29 14 06 4 1

These appear to be the number of patients remaining alive at the start and then at end of each six month period thereafter. The lower numbers are regarded as the DPS stimulation group. Consider the following:

-7 -5 +5 +5 -1

These numbers are the differences in the number of patients who left the study (passed away) at each six month interval. Two things to note:

1. In the six month periods ending at eighteen months and two years the DPS stimulation group had five more survivors than the NIV only group.

2. The -7 in the first six months probably includes the five who opted out and the one found during surgery not to be able to accomodate the DPS device (total of six). They didn't immediately die.

So adjusted, the numbers would be:

-1 -5 +5 +5 -1

. . . far closer than you would think (the DPS stimulation group clearly prevailed in two of the five periods) . . . . Important question is: the stati of those six are (were they still alive at thirty months)? A difference of only one in a period is probably too close to call but two such periods leaned to the NIV only group and in the second period the NIV only group was clearly ahead.

Further, Table 4 lists one DPS stimulation patient as dying of hypothermia . . . correct me if I am wrong, isn't hypothermia freezing to death? How can the DPS device cause one to freeze to death, stimulation or no stimulation?