Going the extra mile? General practitioners' upcoding of fees for home visits.

General practitioners' (GPs') income often relies on self-reported activities and performances. They can therefore 'game the system' to maximize their remuneration. We investigate whether Danish GPs game their travel fees for home visits. Combining administrative and geographical data, we measure the difference between GPs' traveled and billed distances. We exploit a rise in the fees for home visits. If there is a link between the rise in fees and upcoding, we interpret this finding as indicative of gaming behavior. We find that upcoding occurs slightly more often than downcoding (16% vs. 13% of visits) for visits that can be both upcoded and downcoded. Using linear probability models with GP fixed effects, we find that the fee rise is associated with a reduction in upcoding of 0.6% of home visits (2.8% for visits where upcoding is feasible) and no change in downcoding. Importantly, we find no statistically significant differences in the reduction in upcoding across distance bands despite large differences in their fee rises. We therefore conclude that there is no causal evidence of GPs gaming their fees.

Time to retire? A register-based study of GPs' practice style prior to retirement.

In many healthcare systems a large share of general practitioners (GPs) is retiring. The literature has shown a negative correlation between physicians' age and their quality of care. However, little is known about whether GPs exhibit different practice styles in the years prior to retirement. This study investigates whether GPs who are closer to retirement make different professional choices than GPs who are not as close to retirement. Using detailed administrative data on 555 Danish GPs and their patients from 2005 to 2017, we study GPs' practice styles across a ten-year period prior to retirement and compare these with GPs who retire at a later date ('non-retiring GPs'), while controlling for age differences as well as exogenous factors affecting healthcare provision. We focus on the GPs' number of enlisted patients, revenue, provision of consultations, and treatment behaviour in consultations. We find no differences between retiring and non-retiring GPs for key outcomes such as 'revenue per patient' and 'consultations per patient'. However, we find that retiring GPs have fewer enlisted patients in their final years of practicing. This finding is driven by more patients leaving rather than fewer patients joining their lists. We also find that retirement is associated with other dimensions of GPs' practice style, e.g. their provision of home visits, prescribing, and referral rates. Overall, we find a modest association between GPs' retirement and their practice style.

Home sweet home: GPs' response to an increase in the fee size for home visits.

In many health care systems GPs receive fees for their services. Policymakers may use the size of these fees to try to incentivise GPs to provide more care. However, evidence is mixed on whether and how GPs respond to an increase in the fee size. This study investigates how GPs respond to an average increase of 150% in the fee for a high-effort and infrequent service such as a home visit due to patients' illness. We consider Danish GPs' provision of these visits to enlisted patients living outside of nursing homes. Using linear regressions with general practice fixed effects and a rich number of control variables, we estimate the association between GPs' provision of these home visits and the fee rise. On average, we find no association between the fee rise and GPs' provision of home visits. However, we find that GPs who previously provided the fewest home visits to eligible patients increase their provision by 13% after the fee rise compared to other GPs. This increase in visits is driven by more patients receiving multiple visits after the fee rise. We conclude that a fee rise may not yield a strong response in GPs' provision of high-effort and infrequent services such as home visits.