Incentives and interventions in primary care: The effect on service provision, resource use and turnover

Abstract

In many countries, general practitioners (GPs) play a fundamental role in primary healthcare services. Owing to their prominent role in healthcare, the increasing difficulty many Nordic countries experience in recruiting and retaining GPs is alarming. To meet increased patient demand, policymakers must understand how to use their limited resources to increase the supply of GP services. To help advance the evidence base, we investigate how incentives and interventions affects service provision, resource use and turnover amongst GPs. Using experiments, theory, and analysis of large datasets, we show that general practitioners are influenced by economic conditions; both in terms of the services they offer, how much they work, how they code treatments, and whether earnings affect the likelihood of quitting. In Paper 1, we develop a theoretical model to predict how different remuneration schemes and policy measures affect GP effort and the profession’s overall attractiveness. The results show that none of the remuneration schemes increase both the profession’s attractiveness and GPs’ effort levels. We propose that improving working conditions is the only policy unequivocally positive for both these outcomes. In Paper 2, we empirically investigate the sources of earnings differentials and whether these translate into quitting decisions. Using registry data, we find that a 1% increase in income per consultation results in a one-percentage point reduction in the likelihood of quitting. We hypothesise that the sources of earnings differentials are higher effort and better system familiarity. In Article 3, we estimate the effect of an information campaign regarding the appropriate use of a double-consultation fee. We find that the campaign results in a 4–5 percentage point reduction in the use of such fees. In Article 4, we perform a field experiment using a nudging campaign to increase annual checkup attendance for people with type 2 diabetes. We find that sending an emails and SMS leads to a 4% increase in attendance. Overall, our findings suggest that there is no silver bullet; financing systems that induce effort are unattractive to some doctors and more prone to opportunistic behaviour. The changing composition of GPs also reflects different preferences for organisational form, suggesting that a larger suite of contractual arrangements should be offered. Regardless, non-financial interventions will be necessary to ensure timely access to services and maintain collective trust in the system.