> For the complete documentation index, see [llms.txt](https://research.ambitiousimpact.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://research.ambitiousimpact.com/ambitious-impact-eval/suffering-adjusted-days-methodology.md). # Suffering Adjusted Days Methodology Quantifying animal welfare cost-effectiveness under uncertainty. Contributors: Vicky Cox, Sam Hilton, Morgan Fairless Assessing the potential cost-effectiveness and effectiveness of animal welfare interventions requires quantification under wide uncertainty about many inputs, as well as ethics. At AIM, we have quantified animal welfare impacts in different ways over time, attempting to refine our methods. This has included [Welfare Points](https://www.charityentrepreneurship.com/post/is-it-better-to-be-a-wild-rat-or-a-factory-farmed-cow-a-systematic-method-for-comparing-animal-welf) and [Suffering Adjusted Days (SADs)](https://www.charityentrepreneurship.com/post/sads-suffering-adjusted-days-an-improved-animal-welfare-measurement-tool). This document clarifies our current approach to quantifying animal welfare impacts. Our latest changes in early 2026 (which we have called SADs v.2), included: * A decision to primarily report SADs in welfare footprint pain categories * Inclusion of confidence intervals about welfare ranges and sentience in our model * Re-calibration of pain scaling inputs, which we include in our now deprecated reporting of disabling equivalent SADs. ## Suffering-adjusted days and scaled suffering-adjusted days We define Suffering Adjusted Days (SADs) as the > "amount of days in a certain type of pain, conditional on sentience and welfare range." For annoying pain, this is equal to: $$ SAD\_{annoying}= DAYS\_{annoying} \times WRange \times p(Sentience) $$ Scaled SADs are the > "amount of disabling pain equivalent days conditional on sentience and welfare range." This is equal to $$ SAD\_{\text{disabling equivalent}}= WRange \times p(Sentience) \times \sum(WFootprint \times Scaling) $$ Where p(Sentience) is the chance that the animal is sentient; WRange (Welfare Range) is “the difference between the best and worst welfare levels that the individual can realize, where the relevant possibilities are something like the realistic biological possibilities for that individual” [(Fischer and Timmerman 2024, chapter 1)](https://paperpile.com/c/Rw7tnw/7ATx/?locator_label=chapter\&locator=1); WFootprint (Welfare Footprint) is the days in pain by type (annoying, hurtful, disabling, and excruciating) using welfare footprint’s definitions and system; and Scaling is the factor used to convert pain estimates to disabling equivalent pain. ## Current approach Our current approach attempts to quantify uncertainty by outputting 90% confidence intervals based on Monte Carlo simulations for a given scenario and set of issues. Models currently use point estimate inputs for days in a given type of pain, and confidence ranges for welfare range and sentience. Future updates may include uncertainty ranges for days in every type of pain. Scaled SADs involve an extra step, converting all pain estimates into Disabling Pain Equivalents using point estimates of scaling ratios. This is a legacy approach that we keep for information only. In April 2026, we decided to prioritize disaggregated results given our high degree of uncertainty over our pain scaling ratios, and the technical complexity involved in modeling this uncertainty. ### When to use SADs at all SADs introduce a large amount of uncertainty and subjective judgement calls and so shouldn’t be used unless necessary. SADs are an abstract metric that enables a coarse comparison of interventions that help different species of animals. * If you’re comparing two interventions that both help the same species, the same way (e.g. freeing hens from cages) you don’t need SADs: You can compare the number of hens freed from a cage per dollar – this is the most comprehensible and avoids introducing unnecessary uncertainty or judgement calls. If you’re comparing two interventions that both help the same species, in different ways (e.g. freeing hens from cages vs. reducing stocking density on broiler farms) you may still be able to get away without using SADs: You can compare the amount of time spent at each level of pain and not bring p(sentience) or welfare ranges into it. ## How to use the model The SADs system is based on four sheets hosted in a [Drive folder](https://drive.google.com/drive/folders/15u-_YaLh6h7AD2tSO7opNxaro88ChO1Q?usp=sharing) (view only, please make your own copies if making use of the system and external to AIM). * Results and Welfare Footprint Sheet per animal (e.g., [Pigs](https://docs.google.com/spreadsheets/d/1UzWBeMI0wFWa_-i1d1tJSKny9qZ_Q_NEsXqSb8I_xlI/edit?gid=108133649#gid=108133649)) * Feeder sheets that get pulled into the results sheet * [Sentience sheet ](https://docs.google.com/spreadsheets/u/0/d/1poQ2wpGvH_rbov90R0hGAefvQDsidipHPxyhQ0wYfBs/edit)- documenting our process to arrive at sentience estimates for each animal. * [Welfare Range sheet](https://docs.google.com/spreadsheets/d/1-PLsDW1SNXPhNqPel8inQ1Bgp7qFr3m1Ag6MNZ3AR4I/edit?usp=drive_link) - documenting our process to arrive at welfare range estimates for each animal. * [Pain Scaling sheet](https://docs.google.com/spreadsheets/d/1S-SMXcgvqWEFpv63pSeqDbmCfHdN3x5sdduPoS6wn80/edit?usp=drive_link) - documenting our process to arrive at scaling ratios. We detail methodology for each input in a latter annex. Inputs from the feeder sheets are pulled to the animal results sheet, where we document the breakdown of issues as per the welfare footprint methodology, and carry out a monte carlo analysis of results. Our [Shrimp Results](https://docs.google.com/spreadsheets/d/1GaoWLu04bB5MZHqnVLnGCqhLl1uGRe96LV808FNJf0Q/edit?gid=0#gid=0\&range=B15) sheet contains explanatory notes to detail how the sheet works. ## Conventions * Report SADs in whole numbers (decimals can give a false sense of precision) * Report SADs in the following order: P5-P95 (MEAN). * Report SADs without pain scaling as your main result. * You may report scaled SADs alongside separate estimates for each pain category, but note that this introduces highly subjective judgement calls about the relative disvalue of different categories of pain * Document the date of the assessment * Include a full report of your inputs into the model, including a copy of the spreadsheet that led to your calculations (with PULL values hard-coded). * Document any changes made to the estimation approach deviating from the one detailed in this document. [Annex + Google Docs version](https://docs.google.com/document/d/1pCA7TQ8hRvA59DADueU-_yy4Yf9HWzYxWb1ALtb7e-0/edit?usp=sharing) --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://research.ambitiousimpact.com/ambitious-impact-eval/suffering-adjusted-days-methodology.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.