You are reading 1 of 2 free-access articles allowed for 30 days
If you were to ask how many GP visits take place in Ireland annually, you would likely get many different responses.
The answer you receive will have been influenced by several factors, such as the nature of the information contained in the study and the methodology used in gathering that data.
But why is attaining this information so problematic, you may also wonder. Knowledge of accurate GP attendance figures is not only crucial in the context of ongoing and long-overdue GP contract negotiations, but also for the further roll-out of free GP care to other age groups.
It is vital that a consensus is reached on accurate GP attendance rates in order to calculate the impact of any extension of free GP care to ensure necessary resources are in place, for patients and GPs.
In light of all these factors, it would seem plausible to assume that the HSE and/or Department of Health routinely collect and collate such data.
But according to the HSE, it does not gather data on routine daytime consultations. GPs are paid an annual capitation rate per medical card-holder, irrespective of how many times they visit their GP, a HSE spokesperson outlined.
In spite of the lack of HSE data collected, when contacted about GP attendance figures, the Department of Health told the Medical Independent (MI) that as it was a service matter, the query would be more appropriately addressed by the HSE.
However, the Department’s spokesperson pointed out that the Healthy Ireland Survey 2016 included self-reported information on GP visits.
This was an interviewer-administered face-to-face survey, for which 7,498 interviews were conducted, with a representative sample of the population aged 15 and older living in Ireland.
The data has been widely reported and showed that 72 per cent of people aged 15 and over had visited a GP in the previous 12 months, with an average of 4.5 visits per person.
Some 31 per cent had consulted a nurse in a GP surgery in the previous 12 months, with an average of 1.4 visits per person aged 15 years and over. The survey did not calculate a national GP attendance figure, however.
According to Dublin GP Dr William Behan, who has published papers on GP attendance rates, the above data equated to almost 28 million annual GP and nurse consultations.
Using Irish population data, which stated there were 4.7 million people in Ireland (Central Statistics Office, CSO, 2015), the average 4.5 GP visits per person added up to over 21 million GP-only consultations.
The nurse visits equated to 6.7 million nurse-only consultations, leaving a total of almost 28 million GP and nurse consultations, Dr Behan stated.
According to market research company IPSOS MRBI, which conducted the Healthy Ireland Survey 2016 on behalf of the Department of Health, it used four-week recollection and trained interviewers to gather data.
Because four-week recollection was utilised — and not 12-month recollection, as has been used in several other studies in the past — Dr Behan believes the survey provides an accurate figure of GP attendances.
US social scientist and survey expert Mr Charles Cannell identified many decades ago that the number of events forgotten increases proportionately with the length of the recall period
In relation to GP visits, an IPSOS MRBI spokesperson told MI that respondents were specifically asked: ‘How often in the last four weeks did you consult a GP on your own behalf, excluding nurse only-consultations?’
Yet several other studies, including many conducted by the CSO, have used 12-month recollection, a technique widely believed to cause respondents to underestimate their GP visits.
US social scientist and survey expert Mr Charles Cannell identified many decades ago that the number of events forgotten increases proportionately with the length of the recall period.
Mallow GP Dr David Molony, who has also conducted research on GP attendances, told MI that in 2001, the CSO figure for GP attendances was 19 million, but that this reduced dramatically in the mid-2000s, when the recall rate was altered from four weeks to 12 months.
Dr Molony said an explanation on why the change occurred has never been provided.
To confuse matters further, the Irish Health Survey 2015 published by the CSO showed that 74 per cent of people visited a GP in the previous 12 months, with an average of 6.2 annual visits recorded. Four-week recollection was used when interviewees were questioned.
The survey showed 34 per cent visited a nurse only in a GP practice in the last 12 months, with the average number of visits at 2.9.
Using a national population rate of 4.7 million (2015), Dr Behan calculated that 29 million GP visits and 13.6 million nurse visits occurred, which gave a total of around 42 million visits.
It is clear the CSO figure is vastly different to that calculated from the Healthy Ireland Survey. So which one is correct?
Dr Behan believes the Healthy Ireland Survey is more accurate, as the question asked of interviewees in the CSO survey possibly allowed for respondents to include non-clinical activity when recollecting the frequency of visits.
This included, for example, when a patient called a GP surgery to ask for their blood result. Such contact could have been included by respondents in their answers when actual visit numbers were what was being sought, it is understood.
The above example highlights the complexities and nuances involved in attaining accurate figures from self-reported surveys.
Dr Behan argues that many self-reported studies on GP attendances — for example, Growing Up in Ireland (GUI) and The Irish Longitudinal Study on Ageing (TILDA) — are inaccurate, as they use 12-month recollection instead of four-week recollection.
He has analysed data on visits using information provided by GP surgeries and believes this method is an accurate and reliable way of ascertaining GP attendances.
In 2013, an Irish study by Behan et al titled Are Irish Adult General Practice Consultation Rates As Low As Official Records Suggest? A Cross Sectional Study At Six General Practices suggested that 24 million GP consultations occurred annually.
The study of more than 20,000 patients in six GP practices anticipated an increase of 4.4 million consultations a year if free GP care was available to all, with annual consultation rates rising from 24 million to 28.5 million (18 per cent).
The study challenged the accuracy of data from other self-reported studies and suggested that consultation rates were higher than outlined in 2007 and 2010 CSO data.
In 2014, a paper published in the IMJ argued that the introduction of free GP care for children under the age of six could lead to an extra 750,000 GP consultations per year.
Published by Behan et al, the paper was called Does Eliminating Fees at Point of Access Affect Irish General Practice Attendance Rates in the Under 6 Years Old Population? A Cross Sectional Study at Six General Practices.
The report Medical Workforce Planning — Future Demand for General Practitioners 2015-2025 (HSE National Doctors Training and Planning) used data from the CSO, TILDA, GUI and the HSE in estimating GP consultation rates.
The report noted that “in the absence of ‘actual’ and representative GP consultation rates-related data, the consultation rates data presented… are derived from best available information”.
It is understood the report used a Department of Health-advised figure of around 14 million annual GP attendances. But it estimated reasonable increases in attendances with free GP care for all, Dr Behan outlined.
Chair of the IMO GP Committee Dr Pádraig McGarry has also noted the differences in study figures. He told MI: “If you don’t have accurate numbers, how can you project the capacity to deliver these consultations?”
Evidence to date of the impact of the under-sixes roll-out on GP attendances is largely anecdotal, with some surgeries reporting surges of as much as 30 per cent.
But in other practices, perhaps those that have smaller numbers of patients under six, minimal increases have been reported.
What is evident, however, is that the out-of-hours service has experienced large increases in attendances. A quarter of all contacts managed at D-Doc over Christmas were from children under six.
Mr Paul Gorecki at the ESRI last year produced a working paper titled The Impact of Free GP Care on the Utilisation of GP Services in Ireland: An Evaluation of Different Approaches.
In the paper, Mr Gorecki said that if free GP care was available to all, GP visits would increase by 3 per cent and not 18 per cent, as suggested by Dr Behan.
In it, Mr Gorecki seemed to agree with Cannell when he stated that “the evidence suggests that self-reported GP visits tend to underreport actual GP visit rates derived from administrative data”.
Despite this, however, Mr Gorecki favours self-reported studies over analyses of GP surgery attendances.
He believes that where there is a tendency to under-report in self-reported studies like GUI, TILDA and the CSO Quarterly National Household Survey, “this tendency is unlikely to be large”.
When contacted by MI, Mr Gorecki said that in the context of extending free GP care to all, it is important to be able to estimate the likely impact of such a policy initiative on the workload of GPs.
He argued that data on patient characteristics and not just visits are required to do this.
“At the moment, the longitudinal data of TILDA and GUI collects all these data, although concerns have been raised about patient recall in these surveys. Presumably, this could be corrected if survey respondents were asked to keep some sort of record of GP visits,” Mr Gorecki said.
“A survey of GP practices, suitably stratified by size and location to ensure representative sample, sounds like a sensible idea. However, for reasons set out in the paper, it is likely to experience problems in terms of taking into account: Patients who visit more than one GP practice; patients who do not visit a GP practice at all; and defining a visit (eg, telephone, face-to-face, although this applies to GUI and TILDA as well).
“Furthermore, as shown in some of the papers reported in the ESRI paper, problems have arisen with coding in such exercises. GPs typically have limited data on a patient.
“In considering future demand for GP services, it is important to consider what precisely the patient is seeing the GP for and the degree to which other health professionals can substitute for the GP… The collection of such data, together with GP practice characteristics — size, presence of other health professionals, service volume, services offered and so on — would be of considerable use in planning for future primary care policy.”