^'ptembef^gi™'  ^       Crucial  Test  of  Therapeutic  Evidence.  395 
The  report  was  furnished  by  a  physician  for  whom  I  have  a  high 
personal  regard.  I  introduce  it  here,  not  so  much  in  a  spirit  of 
criticism,  but  as  a  justification  of  the  opinion  that  I  have  formed 
of  clinical  evidence  obtained  by  manufacturers  through  their  clinical 
adjutors. 
When  commercial  firms  claim  to  base  their  conclusions  on  clin- 
ical reports,  the  profession  has  a  right  to  expect  that  these  reports 
should  be  submitted  to  competent  and  independent  review.  When 
such  reports  are  kept  secret,  it  is  impossible  for  any  one  to  decide 
what  proportion  of  them  are  trustworthy,  and  what  proportion 
thoughtless,  incompetent  or  accommodating.  However,  if  this  were 
done  it  is  quite  possible  that  such  firms  would  find  much  more  dif- 
ficulty in  obtaining  the  reports.  Those  who  collaborate  should  realize 
frankly  that  under  present  conditions  they  are  collaborating,  not  so 
much  in  determining  the  scientific  value,  but  rather  in  establishing 
the  commercial  value  of  the  article. 
Often  the  best  type  of  clinical  reports — those  in  which  the  ob- 
servations are  directed  to  the  significant  events  and  not  to  mere  side 
lines,  and  in  which  the  significant  events  are  correctly  and  adequately 
reported — generally  lack  one  important  essential,  namely,  an  ade- 
quate control  of  the  natural  course  of  the  disease. 
Since  this  cannot  be  controlled  directly,  it  must  be  compensated 
indirectly.    For  this  purpose,  there  are  available  two  methods : 
The  first  is  the  statistical  method  in  which  alternate  patients 
receive  or  do  not  receive  the  treatment.  This  method  can  usually 
only  be  of  value  when  a  very  large  series  of  patients  is  available. 
Even  then,  its  value  is  limited  or  doubtful,  because  it  cannot  take 
sufficient  account  of  the  individuality  of  cases. 
The  second  method  consists  in  the  attempt  to  distinguish  un- 
known preparations  by  their  effects — the  method  that  might  be 
called  the  "  comparative  method  "  or  the  "  blind  test." 
In  this,  the  patient,  or  a  series  of  patients,  is  given  the  prepara- 
tion which  is  to  be  tested,  and  another  preparation  which  is  inactive, 
and  the  observer  aims  to  distinguish  the  two  preparations  from  their 
effects  on  the  patient.  Surely  if  the  drug  has  any  actions  at  all,  it 
will  be  possible  to  select  correctly  in  a  decided  majority  of  the  ad- 
ministrations. 
The  same  principle  can  be  applied  in  distinguishing  the  superior- 
ity of  one  preparation  over  another.  In  this  case,  the  two  prepara- 
tions would  be  given  alternately  to  different  patients,  and  the  ob- 
