Background: To combat shortages of General Practitioners (GPs) and nurses, the NHS has encouraged the introduction of non-traditional roles in primary care. These non-traditional practitioners, though less taxing on NHS budgets, tend to have less formal training than Doctors, which may affect quality of care.

Aim: To assess the extent to which non-traditional practitioners recognise and meet patients’ needs relative to doctors in primary care.

Methods: We use data from 602,808 respondents to the General Practice Patients’ Survey, a repeated cross-sectional survey randomly distributed to individuals registered at any GP Practice in the UK. Our outcome is a binary variable indicating whether patients felt that their needs were met during their last appointment. Our exposure is represented by seeing a non-traditional practitioner instead of a nurse. We use a recursive bivariate probit approach to account for differences in staffing levels across the practices In our sample that influence the probability of seeing a non-traditional practitioner. We instrument our exposure using the number of staff FTE (per 1,000 patients) who are in medical roles not traditionally associated with primary care, which we argue to be plausibly exogenous from the perspective of patients.

Discussion: Results suggest that patients are less likely to feel that their needs were met when seeing a non-traditional practitioners instead of nurse. Discrepancies could be due to lower ability of non-traditional practitioners to treat patients. Heterogeneity analyses suggests that less crowding of GPs (which may lead to less delegating/more supervision), may lead to better outcomes for patients seeing non-traditional practitioners instead of nurses.

Do English emergency departments respond to demand shocks reordering patients based on non-clinical characteristics? Using plausibly exogenous daily attendance variations, we find small but significant disparities by ethnic background and biological sex. Compared to White patients, Asian patients queue behind one additional every two who unexpectedly join the queue. Similarly, Black patients queue behind two additional patients every five. Differences for females are less striking but significant. These inequalities stem from minority patients failing to be prioritised over those already in the queue. While disparities exist, they do not appear to significantly impact short-term health outcomes.

In most OECD countries, recruitment and retention of General Practitioners (GPs) is proving difficult. Despite this, little is known about observed GP quitting behaviour, and how it relates to earning differentials. Leveraging unique Norwegian registry data for 2010-2019, we explore characteristics associated with GPs’ earnings in Norway and estimate the effect of income per consultation on their decisions to quit. To overcome endogeneity in the relationship between the quitting decision and income per consultation, we use an instrumental variable approach based on plausibly exogenous patient case-mix (specific diagnostic bundles). We find that higher workload is associated with higher average income per consultation. Moreover, conditional on practice characteristics, we find that GPs earning less income per consultation are substantially more likely to quit. Our local average treatment effect (LATE) estimates suggest a 10 percentage points drop in the probability of quitting for a 10 percent increase in average income per consultation.

Background: Healthcare spending covered through mandatory health insurance in Switzerland experienced sustained growth over the last decade. Whilst accounting for a modest share of overall costs, physiotherapy has seen high growth rates in the last couple of years, especially in Southern and Western cantons. Physiotherapy treatment covered by mandatory health insurance is subject to prescriptions issued by general practitioners; the standard treatment schedule consists of a maximum of 9 sessions. Despite systematic gatekeeping, the extent to which Swiss residents use physiotherapy appropriately remains an empirical question. Furthermore, understanding drivers of healthcare growth, and namely identifying the role of incentives on the supply- and demand-side, as well as the gains from care integration in primary care, is crucial for policymaking.
Objectives: This study investigates the burden and drivers of physiotherapy utilisation on the demand- and the supply-side in the setting of the Swiss mandatory health insurance.
Methods: We use data from a large Swiss health insurance company covering about 15% of the population. Firstly, we use a bunching approach to study the distortionary effects of the arbitrary set of 9 visits. Secondly, we exploit changes in exposure to deductibles in a regression discontinuity design framework to unpick the effect of demand-side moral hazard. Thirdly, we derive proxies of practice style for physiotherapists and prescribing GPs. Fourth, we use fixed effects (within) models to explore how practice styles influence physiotherapy demand responses to (lack of) deductibles.
Finally, we use the flow of shared patients between physiotherapists and GPs to study the role of care integration on physiotherapy demand.
Results: Compared to an assumed smooth profile of physiotherapy benefits, the analysis reveals striking anomalies in physiotherapy utilisation related to the standard schedule of 9 treatments. Based on this high-level analysis. Focusing on specific marginal decisions of interest, we find that about 30% physiotherapy sessions can be attributed demand-side insurance-related moral hazard. Furthermore, the study supports the hypothesis of supplier-induced demand. Firstly, moving from physiotherapists with conservative to those with a more aggressive practice styles increase physiotherapy volumes by about 20-30% relative to the average number of sessions. Secondly,
supply-side practice style mediates the likelihood of wasteful use due to demand-side moral hazard. Finally, stronger collaboration between primary care providers and physiotherapists has meaningful effects on utilisation patterns.
Discussion: These findings highlight the significant interaction of both system, demand-side moral
hazard and supply-side factors on health system inefficiency. Policies to curb the growth of
unnecessary spending should be designed accordingly

This commentary (non-research) piece explores the GP Direct Access policy announced in November 2022, which was intended to contribute to this goal, speeding up cancer diagnoses by reducing waiting time for a diagnostic imaging for cases that do not meet National Institute for Health and Care Excellence criteria for an urgent suspected cancer referral. The policy aims to allow GPs in England to directly order CT scans, ultrasounds or brain MRIs for patients with concerning symptoms, but who fall outside the NICE guideline threshold for an urgent suspected cancer referral. Overall, GPs seem to be referring more patients directly to NHS trusts for a cancer investigation since the GP Direct Access policy was introduced. The resulting uptake of direct access investigations varies widely across regions and this might be in part related to resourcing as well as assiduity of regional health organisations in enacting the policy. However, bottlenecks in the system may continue to limit the potential gains from the policy, most likely in a scattered and unequal way across regions in England.

This paper investigates the economic impact of care provider collaboration on healthcare costs in Switzerland. I propose a data-driven approach using Swiss health insurance claims from 2016-2022. I construct patient-sharing networks among primary healthcare providers, applying the Leiden algorithm to identify communities of tight collaboration. For these communities, I derive three key measures: community density, individual provider degree, and the average degree of a main provider’s peers within the same community. I use fixed effects models to study the response to collaboration and connectedness
within-patients, within-providers, and within-communities in terms of patient-levels health expenditure. Preliminary findings reveal that patient and provider responses to average peer collaboration within communities and to community connectedness decrease spending for emergency care and lab tests, but increase slightly spending on medicine. The number of collaborations of a patient’s main primary care provider – a proxy for power – decreases medicines spending. These heterogeneous results highlight the
nuanced role of care collaboration and connectedness in shaping healthcare costs, suggesting avenues for further research into integrated care’s multifaceted economic implications.

We use linked CPRD Aurum data to estimate inequalities in cancer outcomes (stage at diagnosis, mortality within 1 and 5 years) by sex-ethnicity groups and, using the income-deprivation of the patient’s small area of residence, using concentration indices and curves. We then do the same for proxies of primary care phases of the patient pathway prior to cancer: interval between symptoms onset and diagnostic referral, between referral and test, between test and diagnosis, between diagnosis and treatment, staff’s propensity to refer for diagnostic test, GP practice staffing). To estimate the relative contributions towards inequalities in cancer outcomes of patient- and provider-level elements across patient diagnostic pathway in primary care, we use regression-based and Oaxaca-Blinder decomposition approaches.