Julián Victor Méndez Dosuna in Can weakening processes start in initial position? The case of aspiration of /s/ and /f/ relies on Hurch (1988) who, based on yet other scholars, implicates Etruscan into the larger pattern of "/f/-aspiration", just another term for the debuccalization of f to h. At the onset of his examination of the Etruscan evidence, Dosuna offers sound advice:
"Owing to the severe difficulties in interpreting the texts of a language which remains largely unintelligible, one must be extraordinarily cautious in analyzing the Etruscan evidence." (see link)He then delves into some heavily outdated 'evidence' for the alleged f > h sound change in Etruscan:
"It is true that the inscriptions provide numerous instances showing vacillation between the spellingsFirst, Dosuna observes the same p-lenition as I've mentioned many timed before on Paleoglot, minus the conditioning by u. Second, unbeknownst to him, several items he depends on from other academics to be facts are sadly outright fabrications:
and in syllable-initial position: Fasti ~ Hasti (woman's name), safin- > sahin-, fastntru ~ hastntru, faltu > haltu. However, as far as I can judge, Pfiffig's (1969: SS17, 18.2) data seem to indicate that syllable-final /f/ is always secondary and results from the weakening of /p/: cf. hupni ~ hufni, hapna ~ hafna, huplha ~ huf(u)lha. This means that, as in Latin, the distribution of primary /f/ was defective in Etruscan." (see link)
1. There is no *safin- in Etruscan, a word conjectured from an Italic ethnonym.When we clear away the junk linguistics, this alleged Etruscan sound change of f > h rests solely on foreign onomastics. Debuccalization surfaces particularly in the region of Clusium where the variation between f and h is historically most attested. The truth is that Fasti ~ Hasti, Fastntru ~ Hastntru, and Faltu ~ Haltu are all Italic names and have nothing to do with the qualities of the Etruscan language. Absurdly, the gentilicium Ferclite (ET Cl 1.835), attested in Etruscan yet again in Clusium and which is known to stem from Greek Ἡρακλείδης, would seem to show a reverse change of h to f! This can only make sense if the two-way f/h alternation is the result of confusion in the pronunciation of names in a linguistically diverse area, not a genuine sound change in Etruscan itself. Looking beyond Etruscan, in fact, the alternation is likely more illustrative of neighbouring Faliscan, an Italic language that has already shown differences with Latin vis-a-vis the distribution of f and h (eg. Latin hodie 'today' versus Faliscan foied)
2. *huplha ~ *huf(u)lha is surely mistranscribed Θuplθa ~ Θuflθa.
Dosuna later states: "For whatever reason, aspiration did not affect secondary syllable-final /f/." Naturally because it had never occurred in Etruscan at all. Otherwise, for example, we'd expect the Liber Linteus which was written at the close of the 1st millenium BCE to fail to show the many words it does show such as fira (LL 1.xviii) and flanaχ (LL 10.iii), and favin (LL 11.x) which, to the contrary, show that f prevailed with no trace of this sound change.
 Dosuna, Can weakening processes start in initial position? The case of aspiration of /s/ and /f/, Bernhard Hurch y Richard A. Rhodes (eds.), Natural Phonology: The State of the Art. (Papers from the Bern Workshop on Natural Phonology), Berlin, Mouton de Gruyter, 1996, pp. 97-105 (see link).
 Bakkum, The Latin dialect of the Ager Faliscus: 150 years of scholarship (2009), p.209 (see link).
 Read the section entitled The 'f/h' alternation in Faliscan in Stuart-Smith, Phonetics and philology: Sound change in Italic (2004), p.61 (see link).