Waiting for Surgery: Effects on Health and Labor Supply

In universal healthcare systems, patients often face significant wait times for treatment due to constraints on system capacity. While such constraints can serve the purpose of reducing public health spending, the savings are potentially offset if delays in care reduce the efficacy of treatment, increase healthcare utilization, or increase absenteeism from work.

In this paper, we estimate the effects of wait time for orthopedic surgery on healthcare utilization, mortality and health-related labor market outcomes, using microdata covering all publicly funded orthopedic surgeries in Norway referred in 2010-2012. The maximum (mean) wait time in our sample is two years (171 days). Patients are followed for five years from their referral date, which allows us to investigate whether effects persist beyond the usual recuperation period (90 days) for such surgeries.

As the system assigns higher priority to more urgent cases, OLS estimates of wait time effects are likely subject to selection bias. Our identification strategy exploits quasi-random variation in wait times for surgery generated by idiosyncratic variation in system congestion at the time a specific patient enters the queue. Specifically, we instrument for a patient’s wait time using the average wait time of other patients queued for the same procedure at the same hospital around the same time. This instrument strongly predicts individual wait times (F-statistic of 86.8 in main specification) and appears to satisfy the requirement of quasi-random assignment. Our IV estimates should be interpreted as local average treatment effects (LATE) relevant to patients whose wait times are influenced by the degree of system congestion.

We find longer wait times significantly increase healthcare utilization and health-related work absenteeism. Our estimates indicate that, over the five years after referral, hospital inpatient days and the number of general practitioner (GP) visits both increase by 0.4 for every 10 extra days of wait time, while days of health-related absenteeism increases by 7. Small positive mortality effects are also estimated, but are only marginally significant. Back-of-the-envelope calculations suggest the economic costs associated with a patient waiting 10 extra days are 18,500 NOK (approximately $2650) and primarily arise through reductions in work.

Interestingly, the positive effects of wait time on healthcare utilization and health-related absenteeism continue to be evident in years 4 and 5, well past the usual recuperation period for the longest-waiting patients. However, as measured by the occurrence of a repeated surgery, we find no direct evidence that these effects operate through reductions in the efficacy of treatment; the estimated effect of wait time on “resurgery” is negative and insignificant. This would seem to suggest some other mechanism behind the persistence to these effects.