Do patients behave as expected on a healthcare market in transition? Consequences of expanding patient choice and introducing telemedicine in primary healthcare

Abstract

Background: Political governance and technological development have transitioned the Swedish primary healthcare market from a centrally planned, community-oriented model with limited choice for patients to a quasi-market with extensive patient choice. Starting in 2007, reforms promoting patient choice in primary healthcare were gradually introduced in multiple regions in Sweden, and in 2010, all regions across the country were legally required to enable patient choice. Provision of telemedicine – in this thesis defined as healthcare consultations via video calls or text messaging where the patient and the healthcare personnel are spatially separated – emerged in primary healthcare when patient choice was further expanded in 2015. From this point, choice was no longer limited to a certain geographical region, but could extend to primary healthcare providers throughout the country.

Aim: The aim of this thesis was to shed light on the consequences of expanding patient choice and introducing telemedicine in primary healthcare by empirically investigating patient behaviors in response to these changes.

Methods: To approach the aim, I departed from four assumptions about patient behavior which related to four central themes in healthcare: quality, continuity, equity, and demand. The assumptions I investigated were: Quality assumption: Patients are sensitive to quality in their choice of primary healthcare provider. Continuity assumption: Lack of informational continuity leads to worse health outcomes. Equity assumption: Telemedicine consultations are used to a greater extent by individuals with lower healthcare needs. Demand assumption: Use of telemedicine consultations increases the overall consumption of primary healthcare. In all studies of the thesis, comprehensive total population registries were used to investigate patient behaviors. In the quality study, Study I, we investigated whether quality influenced choice of primary healthcare provider by exploring the trade-off between distance to chosen primary healthcare provider and subjective and objective quality measures. We investigated choice of provider in 2017 and included all residents of Region Stockholm who had been in contact with primary healthcare at least once in 2016 or 2017 in the study population (n=1.4 million). The trade-off was studied using a conditional logit analysis. In the continuity study, Study II, logistic regression was applied to investigate if patients who had had their inpatient diagnosis recorded in primary healthcare as well – a measure of informational continuity – filled prescriptions of recommended pharmacotherapies to a greater extent than patients whose diagnosis was not recorded. The study population consisted of 19,072 individuals with an inpatient diagnosis of stroke, transient ischaemic attack, or acute coronary syndrome during the period 2010–2013. In the equity study, Study III, we explored the characteristics of users of telemedicine consultations by investigating the socioeconomic, demographic, and contextual determinants for having made at least one telemedicine consultation in 2018. For comparison, face-to-face physician consultations in primary healthcare were included in the study, as well as digiphysical physician consultations, i.e., telemedicine consultations offered by traditional primary healthcare providers also offering face-to-face visits, and telephone consultations by nurses. All residents of Region Stockholm who were registered with a primary healthcare provider (n=2.0 million) were included in the study population. In the demand study, Study IV, we investigated healthcare consumption following a telemedicine consultation and a face-to-face consultation, respectively. The study population included all individuals who had ≥ 1 physician consultation (telemedicine or face-to-face) during the first half of 2018. Telemedicine users were matched 1:2 to users of face-to-face visits on diagnosis and demographic and socioeconomic variables. Their healthcare consumption, in terms of physician consultations, was compared in an interrupted time series analysis.

Findings: The findings of the thesis lent some support to the four assumptions about patient behavior on the primary healthcare market in transition. In the quality study, the findings suggested that quality had a small impact on choice of provider. However, other aspects – such as the care needs of the individuals registered with a primary healthcare provider – seemed much more important. In the continuity study, the findings revealed a clear positive association between recorded diagnosis and use of recommended pharmacotherapies, indicating that informational continuity is beneficial for patient health. Regarding the equity and demand studies, the findings showed that the use of telemedicine was not equally distributed in the population and that telemedicine users increased their healthcare consumption after a physician consultation more than the matched control group consisting of users of only face-to-face consultations.

Conclusions: The findings suggest that the market mechanism “choice” is not sufficient to ensure quality in primary healthcare. The mechanism might be enforced by better information to patients, but quality should be ensured through other measures. Informational continuity seems beneficial for patients and should be encouraged. The measure “recorded diagnosis” does seem to have a valid role as a follow-up indicator in primary healthcare and the level of diagnosis recording could most likely be improved. Regarding telemedicine, it is important to ensure that the fast technological development does not create a digital divide in healthcare. It will grow increasingly important to ensure that those who need an alternative to a digital channel – whether it is for seeing a physician or booking an appointment – are not deprived of this. In addition, measures to promote access to the groups in the population with a particularly low level of telemedicine consumption should be considered in order to ensure that all groups have access to the health services that best suits their needs. Telemedicine and digital innovations have a lot of potential but are also associated with a risk of overuse. The research highlights the need to promptly investigate the costs and benefits of telemedicine, so that its use can be encouraged when it is most cost-efficient and limited when it is not. The methods applied in this thesis may be useful to monitor the development of the area.