Access to general practitioners in South Australia: a population survey

Most Australians make at least one visit to a general practitioner each year1 and, for many patients, GPs are their major point of access into the health care system. Not being able to access a GP in a timely fashion can be very stressful for patients who are ill, and GP appointment delays have been found to predict lower levels of satisfaction with access to care.2 Internationally, delays in scheduling are routinely documented3,4 and improving access to GP services is a key component of many health care reform agendas.

In the United Kingdom, the National Health Service Plan stated that patients should be able to see a GP within 48 hours,5 and substantial investment has been directed to achieving this goal.6 In the United States, timely access to care is one of the six dimensions of health care quality identified by the Institute of Medicine’s very influential report Crossing the quality chasm,7 and the US Agency for Healthcare Research and Quality argued that improving access to health care would result in reduced morbidity and mortality and also lead to cost savings.8

Patient access to general practice appointments is not routinely measured and, in Australia, very little evidence is available to inform discussions on this important dimension determining quality of care. A health population survey in New South Wales found that, of people who experienced difficulties in getting health care when they needed it, the most prevalent difficulty reported was the waiting time for a GP appointment (36%).9 This compares with 14% of people reporting difficulties with access to specialists and 7% with emergency department waiting times. However, the NSW survey did not measure actual appointment waiting times.

Australia is included in the Commonwealth Fund International Health Policy Survey and, in the most recent survey,10 respondents were asked how quickly they could get an appointment to see a doctor the last time they were sick or needed medical attention. In Australia, 62% reported that they were able to be seen the same or next day, 21% waited 2–3 days, and 15% waited 4 or more days. However, the survey did not examine the reasons why people waited for an appointment (eg, if they waited to see the GP of their choice or if an earlier appointment was not convenient for them).

In 2007, Catholic Health Australia commissioned a telephone survey of 1200 Australians aged 18 years and over to elicit perceptions about ease of access to health care services.11 Most respondents (87%) indicated that they believed it was easy to gain access to GPs, but only 50% agreed that it was easy to gain access to medical specialists. Those in capital cities and those in white-collar occupations were more likely to report easy access to GPs compared with those residing in rural areas or employed in blue-collar occupations.11 The results from this survey (in contrast to those from the NSW survey) appear to indicate that most Australians experience good ease of access to GPs.

The purpose of our study was to measure the actual timeliness of access to GPs in South Australia. This was measured by including a number of questions about access to GPs in a state-level population health survey. We also examined possible associations between demographic and other contextual factors (eg, area of residence) and access to GPs.

The data for this study were collected in the 2007 South Australian Health Omnibus Survey,12 a representative survey of people aged 15 years or older. The Health Omnibus Survey uses face-to-face interviews, with a non-replacement sample of households throughout the state. In rural areas, only residents in towns with a population of 1000 or more are sampled.

Respondents were asked a series of questions about access to GPs based on patient satisfaction surveys carried out in England by the Picker Institute on behalf of the UK Healthcare Commission.13 The questions about access to GPs that we used in our study are shown in Box 1.

A key indicator used by the Healthcare Commission to monitor access to GP appointments is the proportion of appointments that involve a wait of greater than 2 working days because the person could not get an earlier appointment. This indicator was constructed to exclude routine appointments, such as those that are made to obtain prescriptions and referrals, and planned appointments required for periodic monitoring of health conditions. It also excludes patients who waited more than 2 days because: they wanted to see a GP of their choice, because it wasn’t convenient for them to be seen earlier, or for other reasons (eg, waiting for test results).

In our study, we used a similar indicator labelled “undesirable waiting time”, which was defined as a waiting time for an appointment of more than 2 days because an earlier appointment was not available. This was calculated from a combination of responses to Question 2 and Question 3 (Box 1). The denominator used was the number of respondents endorsing Question 2, options 2, 3 and 4. The numerator was the number of people waiting more than 2 working days (Question 2, option 4) and who also indicated that the main reason they waited was because they could not get an earlier appointment (Question 3, option 2).

Ethics approval for the survey questions was provided by the South Australian Department of Health – Human Research Ethics Committee (HREC).

Stata/MP, version 10.0 (StataCorp, College Station, Tex, USA) was used to perform the analyses. The analyses take into account that the survey data were weighted to the 2006 Australian Bureau of Statistics census figures by sex, age and demographic variables (country of birth, marital status, educational attainment and household income) to be representative of the South Australian population. Small discrepancies in some table frequency and percentage totals occur due to the effect of rounding in the presence of weighted data.

The 2007 South Australian Health Omnibus Survey had 2507 respondents (participation rate, 62.7%).12 Of the 2507 survey respondents, 2175 (87%) reported visiting a GP in the previous 12 months (Box 2).

The data obtained in this representative population study indicate that, generally, people have timely access to GP appointments in SA. Seventy-eight per cent of respondents reported that they were seen by the doctor as soon as they thought necessary. However, 11% indicated that they should have been seen “a bit sooner”, and 10% indicated that they should have been seen “a lot sooner”. By the UK National Health Service standard for timeliness of access — that consumers seeking care should be seen within 2 working days — we found that that this was not the case for 9% of South Australians, who waited longer because no earlier appointment had been available.

This level of undesirable waiting time, as we have defined it in our article, was lower than that reported by the UK Healthcare Commission. In 2006, 12% of English patients (of 10 000 questionnaire respondents) who waited for an appointment reported that they waited more than 2 working days because no earlier appointment had been available with any GP at their local practice.14 A similar proportion of English respondents (compared with the South Australian respondents in our study; 77% v 78%) reported that they were seen by the GP as soon as they thought necessary.

These results suggest that actual waiting times for GP appointments and patient perceptions of waiting times are slightly better in SA than in the UK. However, these modest differences should not be over-interpreted because small variations in methods between surveys can have large impacts on estimates. An example in this case is that the UK survey was mailed to patients of registered GPs’ patients while our survey was conducted face-to-face. The results for actual waiting times for GP appointments are, however, consistent with those reported by an international health policy survey conducted by The Commonwealth Fund in 2007, which found that Australians were able to see a doctor more quickly than respondents in the US, UK and Canada, but not as quickly as those in the Netherlands and New Zealand.10

The univariate and multivariate logistic regression analyses of demographic and contextual factors associated with undesirable waiting times showed that respondents reporting lower household incomes were at increased risk of experiencing an undesirable appointment waiting time. Living in a regional (versus a metropolitan) area was significant in the univariate analyses, but not when household income was controlled for. Given GP workforce shortages in country areas,14 the finding that South Australians living in regional areas do not appear to be at increased risk of experiencing an “undesirable waiting time” is surprising. However, only country towns with a population of more than 1000 were surveyed, and we are not confident that these results could be generalised to smaller rural communities. Thus, it is important to be clear that our results indicate only that we failed to detect a difference in access to GPs between South Australians living in metropolitan Adelaide compared with respondents living in South Australian regional centres.

The finding that respondents reporting lower incomes were at increased risk of an undesirable waiting time in obtaining a GP appointment adds to growing Australian literature documenting socioeconomic health inequalities. Because access to primary health care in Australia is relatively universal, it is argued that the health care system is not likely to be a major cause of socioeconomic health inequalities.15 Our results suggest that universal access is not necessarily the same as equal access.

The definition of an undesirable waiting time that we adopted is a conservative one because it does not classify waiting more than 2 working days to see the GP of patients’ choice as an undesirable waiting time. If we had classified such appointment waiting times as “undesirable” the rate of undesirable waiting times would have doubled to 18%. In 1999 the Consumers’ Health Forum of Australia released a document indicating that the timeframe for arranging a consultation with a GP for non-urgent cases should be less than 48 hours.16 If this means a consultation with the GP of the patient’s choice, then this is a very high standard and, as our results show, it was not reached nearly a fifth of the time.

Given the high value many patients place on being able to see the GP of their choice, for many health problems, this may take precedence over quicker access. As working hours vary for individual GPs, there will always be some degree of trade-off between choice and availability, and patients of some popular GPs with limited working hours will always have long periods of waiting for appointments, even in settings where timely access to GPs is very good.

No explicit standard has been created for time waiting to be seen in GPs’ surgeries, but in general, waiting times of more than 30–45 minutes are regarded as excessive. We found that waiting times within GPs’ surgeries were quite variable, with a significant minority of patients (13%) waiting 45 minutes or longer. Long surgery waiting times have been found to be associated with poorer patient satisfaction, but it appears that patients are prepared to give GPs considerable leeway in waiting times, provided that they receive adequate time with the GP to discuss their problems.17 Waiting times in GP surgeries are influenced by many factors, including patients presenting with complicated medical problems that had not been anticipated in advance, and the methods used for scheduling availability of GPs. Improved scheduling practices can have an important influence on surgery waiting times.18,19

Several limitations to our study need to be acknowledged. The survey participation rate (62.7%) is reasonable for surveys of this type, but it is possible that people with reduced levels of access may have been less likely to take part in the survey. There is considerable variation in the distribution of the GP workforce across the country, and our results may not generalise to other Australian states and territories. Our broad classification of area as either metropolitan or regional is less than ideal because it probably masks considerable variability; also, smaller rural communities were not included in the survey. Further geographic analysis of the survey data using a finer grain approach to “area” is planned.

The concept of “health access” is complex20 and its measurement challenging.21 Internationally, and particularly in the UK, considerable progress has been made towards better understanding of the role that timeliness of GP appointments and other factors, such as the ability to make after-hours GP appointments, have in determining patient satisfaction with their primary health care system. If Australian policymakers seek to tackle perceived problems with access to GPs, they need to understand what the problems are, where they occur, and their magnitude. Surveys like ours can contribute to this understanding.