What the work actually is

You build economic evaluations end to end — a research question, a defensible model structure, the parameter set and its sources, the analysis, and a recommendation a payer could act on. Then you write the reference answer: not just the ICER or the net budget impact, but the methodological reasoning behind each choice you made along the way (why a partitioned survival model rather than a three-state Markov, why that discount rate, why a societal perspective was or wasn't appropriate). A second stream is grading: you read model-generated analyses and score them for analytical soundness, and you flag the ones that read fluently but would mislead a payer or an HTA reviewer — extrapolated survival curves with no hazard justification, utilities lifted from a mismatched population, a half-cycle correction quietly omitted, cost inputs from the wrong jurisdiction or year.

What the screen is looking for

AfterQuery's screening is AI-led and follow-up heavy. It is checking that you have personally built models rather than reviewed them — expect to be asked about structure choice, cycle length, parameter sourcing, and how you ran and interpreted PSA. It is also checking evaluation judgment, which is a distinct skill: can you separate a wrong number from a wrong method, can you articulate why an assumption is indefensible rather than merely unusual, and can you write that reasoning down so a non-economist reviewer can follow it. Concrete detail from real work — a therapeutic area, a perspective, a data source, a decision you'd defend to a committee — moves the screen forward. Generic method summaries do not.

Logistics

  • Fully remote, fully async. No client deadlines, no billable-hour targets, no meetings you must attend live.
  • Minimum commitment of 10 hours per week; you set the hours and they can vary.
  • Paid hourly via Stripe. Observed band is $90–150/hr, positioned by depth of modelling and HTA experience — treat it as observed, not guaranteed.
  • Ongoing project work rather than a fixed-term engagement.

Useful but not required: prior submissions to NICE, ICER, CADTH, PBAC, IQWiG or AMCP; hands-on modelling in R, TreeAge, Excel/VBA, Stata or SAS; work with MarketScan, Optum, IQVIA, Medicare, SEER-Medicare or Flatiron; ISPOR involvement or peer-reviewed publications applying economic evaluation methods.