Digital Transformation in Swedish Primary Care: Evaluating the Impact on Antibiotic Prescription and Equity in Utilization

Jens Wilkens, Björn Ekman, Hans Thulesius

Keywords: Digital Primary Care, Antibiotic Prescription, Equality and Equity, Utilization

Background:

In Swedish primary care, purpose built digital providers started offering consultations nation-wide in 2016. These services increased fast in volume, driven by low access to primary care and a favourable reimbursement system. The emerging digital landscape of health care services raised concerns about its effects on several key performance dimensions, including quality and equity.

Research questions:

We aimed to assess defined effects of digital primary care consultations in Sweden. The research questions were: Who utilizes digital services; What conditions are they used for; What are the effects on antibiotic prescription and equity in utilization across the population, in comparison with in-office consultations.

Method:

We used a purposively built database with individual level service utilization and linked socio-demographic data. Propensity score design was used to analyse differences in antibiotic prescription to manage the non-random data collection, and regression-based concentration indexes with adjustment for need factors for the equity analysis.

Results:

Mild infections among middle age women with high incomes were relatively more common in digital than in in-office primary care consultations. Contrary to fears expressed among medical professionals, antibiotic prescriptions were less common in digital consultations, also in models that aspire to incorporate effects of treatment selection bias in the data. Digital service utilization was “pro-rich”, also when adjusting for need factors such as age, sex and illness. This contrasted in-office services, which showed a “pro-poor” pattern, just like similar studies of traditional primary care. However, within the patient group using digital primary care, utilization was proportionally distributed across income.

Conclusions:

The studies in this project contribute to the evidence base for further deployment of digital primary care services. Advances in digital services can make use of this knowledge to design payment systems and services, to better target specific conditions and populations which are currently underserved.

Points for discussion:

How can traditional primary care services integrate with digital services so that the two forms of contact complement each other?

Which disease groups are likely to grow in digital primary care?

What can clinicians do to mitigate the inequities seen in this study?

#18