Background and purpose of the study:
Offering the right balance of access routes is difficult for GP practices, as patient needs range from urgent conditions to non-urgent services, routine check-ups and follow-up visits. Some practices offer a high level of telephone consultations, whereas, others limit their GP and nurse time to face-to-face appointments.
Whether the consultation is with the patients regular GP or not can also affect the need for the patient to return to primary care. Continuity of care is shown to have many benefits, though there has been a decline in continuity of care in recent years and the exact causes of the decline have been a matter of speculation.
An appointment with the regular GP entails continuity of care as the GP will have information about the patients health status, enabling a more meaningful consultation. We therefore expect the balance between the two types of consultations to vary depending on whether the consultation was with the patients regular or another GP in the practice.
We are interested in answering two main questions:
1. What factors within the primary care setting have driven the decline in continuity of care, and specifically, which factor is most important? Only once we understand these factors, can the GP practices manage continuity better.
2. How does the variation in type of consultations and whether the consultation is with the regular GP or not, affect the primary care and secondary care usage of their patients?
If a practice offers a high percentage of telephone consultations, it may lead to increased A&E activity as patients will use A&E as an alternative access route. On the contrary, a high percentage of telephone consultations may provide easy access to healthcare professionals, which can lead to decreased A&E activity. Limiting face-to-face consultations may lead to more outpatient activity, as GPs increase referrals to manage their limited capacity. Furthermore, these effects can be moderated by seeing the patients regular GP for the consultation.
We aim to understand this relationship between access to healthcare professionals in the GP setting and their patients usage of primary and secondary care resources.
Potential importance of the findings: The NHS General Practice Forward View and NHS Long Term Plan commit significant new resources to GP practices. In return, the NHS expects that GP practices will change their offering to reduce system usage. This research project provides insights into the potential to achieve this through appropriate triaging, balanced appointment offerings and the patients ability to see their usual doctor in GP practices.
Research Question 1:
Study aim and objectives.
We will study and quantify the relationship between different demand and supply factors as exposure variables, and the ability of the patient to see his regular GP as the main outcome variable. We expect that a combination of increase in demand and a reduction in supply leads to a reduction in continuity of care.
Design.
Our design is a longitudinal study that estimates effects of the exposure variable on the outcome variable within practices over time.
Statistical methodology.
We estimate this relationship between exposure and outcome variables by first breaking down the demand and supply factors into relevant and meaningful measures, aggregating the measures at the practice-month level and ultimately estimating a longitudinal model, specifically, a fixed-effects panel data model.
Research Question 2:
Study aim and objectives.
We will study the relationship between a practices balance of telephone to face-to-face consultations (ratio of telephone consultations to face-to-face consultations) as the primary exposure variable, and, patients return frequency to the GP practice(number of face to face consultations in the next year), A&E visits, emergency admissions, emergency bed days and outpatient visits (annualized per capita) of registered practice patients as outcome variables. We will use the percentage of consultations with a patients regular GP as a moderator variable to see if relationship between our outcome and primary exposure variables differs.
We expect that a lower rate of face to face consultations in the current year may increase the patients return frequency to the practice in the next year. We also expect that a higher rate of face-to-face consultations leads to decreased outpatient activity, but a higher rate of telephone consultations at the expense of face-to-face time, may or may not lead to decreased A&E activity. Since capacity is limited, the tradeoff between telephone and face-to-face time for different patient segments, provided by different healthcare professionals, needs to be well understood. These effects may be more or less pronounced if the practice offers a higher ability for patients to have consultations with their regular GP.
We aim to estimate the relationship between exposure and outcomes and to identify a range of telephone appointment rates that balance the effects on A&E and outpatient activity and lead to low overall secondary care usage. We will break this down by the type of healthcare professional as well as patient segments.
Design. Our design is a hybrid design, estimating cross-sectional and longitudinal effects simultaneously.
Statistical methodology.
We estimate the relationship between exposure and outcomes using the following steps. First, we develop prediction models that allows us to predict the annual expected value of the 5 outcome variables as a function of patient characteristics alone. Second, we calculate for each outcome, a standardized outcome for each practice and year as the ratio of the sum of observed outcomes to the sum of model-predicted outcomes. A GP practice with a ratio below 1 will use fewer secondary care resources than expected for its patient population, a practice with a ratio above 1 will use more. Third, we estimate the relationship between exposure, standardized outcomes and the moderator, using multi-level longitudinal models. We use subsample and interaction models to assess the effects on subpopulations.
Primary outcomes of interest throughout the study: per capita annual number of A&E visits for the practice population; per capita annual number of emergency admissions for the practice population; per capita annual number of emergency bed days for the practice population; per capita annual number of outpatient visits for the practice population.
The capitation (size of registered list) of a practice will be estimated from the observed number of patient consultations in a year, using published average numbers of annual GP patient consultations by age groups (1).
For reporting purposes, we would only specify practices as small/medium/large.
Stefan Scholtes - Chief Investigator - University of Cambridge
Stefan Scholtes - Corresponding Applicant - University of Cambridge
- Collaborator -
- Collaborator -
Harshita Kajaria - Collaborator - University of Cambridge
Michael Freeman - Collaborator - INSEAD
2011 Rural-Urban Classification at LSOA level;HES Accident and Emergency;HES Admitted Patient Care;HES Outpatient;Patient Level Index of Multiple Deprivation