The total rate of energy use in the united states is now such that each person has the equivalent of about horsepower working continuously.
98.3 qbtu/year
29 million people in the u. s.
horsepower= 746 w
the above is the info you will need. this is what i have done so far, but it can not be right:
98.3*10^15btu/year (1 year/365 days) = 0.269315*10^15 btu/day (0.269315*10^15 btu/day)/291 million = (0.2693*10^15 btu/day)/291*10^6 ~ 0.00092548109*10^5 btu/day = 925,481.09 btu/day
925,481.09 btu/day*(1day/24hour) = 38561.71 btu/hour ~ 38561.7 btu/hour
1 horsepower = 746 w
1 w = 3.412btu/hr 38561.7 btu/hr*(1 w/3.412 btu/hr) ~ 11301.3 w*(1 hp/746 w) = 15.149 hp

The more I have worked on this, the stranger it has become.
First ... I didn't know what 'QBTU' means.  But I see that you called it
' 10¹⁵ ' so I guess it's 'quadrillion'.
Next ... your given data says 29 million people in the US, and that's
ridiculous. Right now it's about 319 million.  It was 290 million in 2002,
and I see that you used 291 million, so that looks like it was a typo. 
No problem.

Now . . . I did it 3 or 4 times using that 3.412 BTU/hr = 1 watt .  I did get
the significant figures of 152, but I kept getting either 152 HP or 0.152 HP.

At this point, I traced through your solution ... thank you very much for
posting it ... and I'm going to take your colossal load of points for saying
that I I did go through every step in detail, I agree 100% with everything
I see there, and I endorse every move you made.    

So up to now, we both agree that we cannot see where 1.5 HP/person
comes from.

I looked back to the conversion factors, and I saw something that could
make the arithmetic less complex:  1 BTU = 1,055 Joules
Look what I can do with that:

   (98.3 x 10¹⁵ BTU/yr) x (1,055 joule/BTU) x
               (yr/365 day) x (day/86,400sec) x
                                     (HP/746 joule-sec) = 4.4082 x 10⁹ HP

then . . .

(4.4082 x 10⁹ HP) / (29.1 x 10⁷ people) = 15.15 HP/person

I'd say that your work, using the given data, has been vindicated by
an outside, independent consultant.  It may not be a true statistic, but
your math is bullet-proof, and the data have been properly implemented.