I wanted to add a more recent review that adds rules citations for three important pieces of information mentioned by earlier reviews/comments. Let's read the fine print on these Hospital Debts!

First: If you draw Hospital Debts naturally as part of the Upkeep phase at the end of the round, you will not be able to sneak two resources onto it before the round ends. The rules for timing make clear that there is no ~Player Window after you draw cards.

Second: In multiplayer, other investigators at the same location can activate and pay for Hospital Debts. In the rules, Activate) Actions can be taken on "all encounter cards in the threat area of any investigator at that location," and according to the rules for Weaknesses,) any weakness with an encounter cardtype (like "Treachery") "are considered to be encounter cards [...] once they have entered play."

Third: In multiplayer, the "Limit twice per round" applies to each investigator individually, meaning that three different investigators could pay down the debts in a single turn. The relevant) rule on limits specifies "Unless stated otherwise, limits are player specific." And this is likely relevant to a number of other cards.

Again, these three points have been mentioned elsewhere, but I wanted to collect them all in one place with relevant citations. I so often find that discovering the rules behind one card makes me better understand other cards, and the game as a whole!

As mitigations for failure go, Self-Sacrifice doesn't quite stand up to Survivor staples such as Take Heart. Take Heart grants a failing investiagtor two resources as well as two cards and can be committed to your own skill tests, so it has great synergy with the Survivor card pool in general.

Self-Sacrifice's specialty comes from allowing you to pass around the consequences for skill test failure. Many punishments for failure include taking damage and/or horror, and I know I've lost a handful of scenarios due to one player drawing and subsequently failing every Rotting Remains in the deck, while another sat pretty with a pristine sanity pool! Self-Sacrifice may allow you to avoid that case in a way that Take Heart won't while providing at least a portion of the older card's tempo acceleration. You can even choose to take the extra draw yourself, which might be good for catching up if you just tanked a chunk of horror due to an auto-fail.

Notably, Self-Sacrifice might also let you tank a Retaliate or Alert attack trigged by a failed test, which may well be relevant if you're packing some of the Guardian cards that reward you for taking attacks, particularly those spoiled for the Nathaniel Cho starter deck (Counterpunch, Lesson Learned, etc.). Weirdly though, this demands that you play someone with a Guardian card pool while your teammate attempts to fight or evade a monster, rather than you taking that action, which doesn't come up often in a typical two player Guardian monster hunter plus Seeker clue gatherer pair.

Clearly this isn't a card for every investigator. Roland Banks tends to be the very investigator suffering from one too many Rotting Remains with his paltry 5 sanity, and the same could be said for "Skids" O'Toole and Tony Morgan with their pitiful willpower. Investigators with larger sanity pools and higher willpower are more likely to appreciate Self-Sacrifice's ability to divert punishment for failure to themselves. I think one key investigator who will appreciate this is Carolyn Fern who is happy to tank horror all day, as its easier to heal on herself than on teammates, and she has an adequate health pool too. Diana Stanley can really ratchet up protecting her teammates with this as well, and she might be able to cancel a consequence for failure with Deny Existence when it's applied to her rather than a teammate.

Interestingly, there are a handful of investigators who have both Take Heart and Self-Sacrifice as part of their card pool. At time of writing they are William Yorick, Lola Hayes, Calvin Wright, and Tommy Muldoon. As per my brief analysis earlier, I think Take Heart is generally the stronger of the two cards because granting extra resources is more straightforward to take advantage of than moving consequences for failure, but some of these investigators might consider Self-Sacrifice in addition to Take Heart. Yorick, Calvin, and Tommy can all benefit from tanking extra damage or horror on their cards: Yorick and Tommy can both replay soak assets and Calvin can boost his stats. Lola is of course a toss-up because of her bizarre deckbuilding restrictions, but I think even if you are playing Guardian as one of your roles, you're too fragile to really consider Self-Sacrifice.

In short, any investigator who is reasonably durable and doesn't mind tanking some treachy failures, damage, and horror ought to consider Self-Sacrifice to protect more vulnerable teammates. The tempo reward of drawing cards actionlessly is a nice bonus for a faction without much reliable and repeatable card draw. It's no staple, but it is a solid card.

This card is certainly better than suggested by the other review, and far from "the worst card in the game" as suggested in the review for the upgraded version. I challenge you to actually use this in an appropriate deck (such as Tony Morgan or Jenny Barnes) and not see his value. I did and I would choose him again. The point is not only extra actions, but also optimizing when you play each item, so that you can maximize the skill value, clues, and damage from your weapons and tools to match what you are facing. He also makes Colt Vest Pocket and Lupara decent weapons, especially with Tony Morgan. Yes, he is expensive, so don't use him with "Skids" O'Toole or other investigators that tend to be short on resources. But in the right deck, the effect is worth the cost. Also, the theme fits perfectly: he is the guy that can get you whatever you need, right when you need it, with no limits, as long as you can pay the price.

I feel this card is most useful for Sefina Rousseau. Sefina has a Willpower of 4 and is naturally event-focused. She's likely to see this early as well thanks to 3 copies from Myriad and a 13 card opening hand. Lockpicks was already fairly strong for her, allowing her to investigate at a base of 6. Lockpicks + Mind's Eye allows her to investigate at a base of 8 (or 9 with Holy Rosary), making it much more likely you can hold onto those Lockpicks supplies at high-shroud location.

While it does compete with Double, Double and Suggestion for your Arcane slots, it's significantly cheaper in exp than the former, and gaining 3 copies on purchase makes it a more consistent find than the latter. Consider taking Mind's Eye as an early purchase while saving towards the 8 XP for Double, Double, then afterwards grabbing a Versatile to throw in your favorite off-class event, a Sign Magick or two, and more of those fantastic rogue events (alternately, some Guts or Lockpicks if you want to mitigate some of the consistency penalty that comes with Versatile's extra cards)

Benefits gained from skill cards are always nice because you never have to commit actions or resources to gain the desired outcome. Cards like Fearless for healing horror, or Quick Thinking for gaining actions. But in theory Nimble potentially provides a whopping 3 free actions! That's insane value, and you're not even paying anything (except the card itself).

So what are the downsides? Firstly it's only valuable if you need the moves. Maybe you still need to gather clues on your current location, or you need to stick around to act as bodyguard. If you don't need to move, then Manual Dexterity is better in every way. Secondly, can you move? The scenario itself could be preventing you from progressing further.

Maybe you'll put it to good use when you evade spoiler and subsequently use the move actions to reach the exit in spoiler. But if you have access to Nimble then you almost certainly have access to Elusive (a fast disengage and teleport), so even the above hypothetical is questionable at best. A very circumstantial card. I predict this card will be most frequently used for a single move action, in which case Quick Thinking is arguably better.

TL;DR - Hard to get maximum value from it, most likely used for 1 free move.