i wonder what sort of dopamine replacement therapy (DRT) they were on, and whether therapies were of the same nature and dose across the two PD groups.

Does anyone have a copy of the “Functional Abnormalities” paper?

As a point of interest, the 2nd study (Dodd 2005) that claimed to have found an association between not just DRT and pathological gambling (PG), but specifically dopamine agonists (DAs) and more specifically mirapex, explicitly exonerated levodopa, in spite of the fact that nearly everyone in that study (and most of the other studies i have read making that claim) was also taking levodopa.

Moreover, the fact that the authors of that study blamed mirapex in particular and dopamine agonists in general didn't stop them from citing, in support of their conclusion, studies concluding nothing of the sort.

the actual conclusions of the cited studies are as follows:
----------------
Giovannoni
Hedonistic homeostatic dysregulation is a neuropsychological behavioural disorder associated with substance misuse and addiction (meds not specified)

Seedat
pergolide (this paper describes a single person)

Geschwandtner
1) predisposition to gambling triggered/exacerbated by overuse of dopaminergic drugs (including levodopa), or 2) lack of dopamine, I.e. PD itself, responsible for gambling

Stacy
chronic high-dose use of dopamine agonists

Montestruc
bromocriptine (this paper describes a single person)

Kurlan
PD/dopamine deficiency itself

Avanzi
Abuse of dopaminergic drugs (including levodopa)
--------------------

So, the tally is:

Abuse of dopaminergic drugs in general –2.5
PD (dopamine deficiency) itself – 1.5
Pergolide – 1
Bromocriptine – 1
High dose dopamine agonist therapy (mirapax in particular) – 1

So of the seven studies cited, only three support a DAs in general conclusion, and only one supports a mirapex conclusion.