Family Practice Vol. 19, No. 1, 85-92
© Oxford University Press 2002
Original Paper |
General practice-specific care categories: a method to examine the impact of morbidity on general practice workload
Greater Murray Clinical School, University of New South Wales, PO Box 5695, Wagga Wagga, NSW 2650, Australia.
Sturmberg JP. General practice-specific care categories: a method to examine the impact of morbidity on general practice workload. Family Practice 2002; 19: 85–92.
Received 5 January 2001; Revised 20 June 2001; Accepted 3 September 2001.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodResultsDiscussionConclusionsReferences |
|---|
Background. Governments are increasing pressure on GPs to provide better services to their patients without giving consideration or due recognition to the impact of those initiatives on their already heavy workload.
Objective. This pilot study aimed to measure accurately the impact of case mix on general practice workload.
Method. The general practice-specific care category (GP-SCC) model was developed and applied to a random sample of patients who attended a four-doctor suburban practice four or more times between July 1995 and June 1997.
Results. The random sample comprised 245 patients (126 males, 119 females) out of a total practice population of 
4000. The mean patient age was 42.7 years (CI 39.6–45.8; range: 0–95). The mean patient consulted 10.70 times (CI 9.62–11.77) and discussed 13.19 health problems (CI 11.74–14.63), which equated to 1.20 problems per consultation (CI 1.17–1.23). The ambulatory case mix concept allowed the development of the GP-SCC model—defined as GP-SCC 1, acute/self-limiting problems and preventive care; GP-SCC 2, primarily chronic health problems; GP-SCC 3, psychological problems in conjunction with up to two other problem categories; and GP-SCC 4, a combination of four or more problem categories. GP-SCC 1 comprised 31.1% of patients (CI 29.1–35.1), accounting for 25.6% of visits (CI 24.0–27.3) and 21.9% of all problems encountered (CI 20.5–23.3); GP-SCC 2 comprised 16.7% of patients (CI 10.6–19.6), accounting for 10.6% of visits (CI 9.5–11.9) and 9.9% of all problems encountered (CI 8.9–11.0); GP-SCC 3 comprised 7.1% of patients (CI 4.4–11.2), accounting for 7.8% of visits (CI 6.8–8.9) and 7.7% of all problems encountered (CI 6.8–8.7); and GP-SCC 4 comprised 42.0% of all patients (CI 35.8–48.2), accounting for 56.0% of all visits (CI 54.2–57.8) and 60.5% of all problems encountered (CI 58.8–62.2).
Conclusions. The GP-SCC model, built on the ambulatory case mix concept, is a useful tool to analyse the morbidity of practice populations, and has a good predictive value in terms of a practice' workload.
Keywords. Ambulatory care, case mix, family medicine, general practice, workload.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionsReferences |
|---|
In recent years, the Australian Government has increased pressure on GPs to provide broader and more comprehensive care with the aim of achieving better health outcomes. Some of these initiatives include ‘quality use of medicine activities’,1 health care examinations for the elderly and case conferencing,2 and health care planning/co-ordinated care trials.3 Expressed in economic terms, the Government is looking to achieve an increased return for their investment into health. Few would take issue with these aims. The problem is that, to date, there has been a failure to take into account the way in which these changes affect the already limited resource ‘provider’.
This pilot study aimed to measure accurately the impact of case mix on general practice workload.
|
Method |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionsReferences |
|---|
The study was conducted in a four-doctor suburban practice on the New South Wales Central Coast. The estimated practice size was 4000, with the doctors providing
22 000 consultations per year. A random number table was used to identify the study population. To be eligible for inclusion in the study, patients were required to attend for at least four consultations between July 1995 and June 1997. The ambulatory case mix concept—previously described by Weiner4 and Starfield5 in 1991—was applied to the study population. For each patient, the medical record was reviewed and all problems encountered over the 2-year period were counted and categorized into diagnostic groups (CADG). The pattern of CADGs for each patient was then collapsed further into major ambulatory categories (MACs).
Data were analysed using the SPSS 10.06 statistical software package. The measures were reported as means with 95% confidence intervals (CIs), while statistical significance of continuous measures was tested by a two-tailed Student t-test, and categorical data were tested for significance by the chi-square test where appropriate. Results were considered statistically significant at a level of 5%.
|
Results |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionsReferences |
|---|
The random sample comprised 245 patients (126 males, 119 females) out of a total practice population of
4000. The mean population age was 42.7 years (CI 39.6–45.8; range: 0–95). In a previous analysis, we showed that the sample characteristics were representative of the practice population7,8 and were also consistent with the population characteristics of the Australian Morbidity and Treatment Survey (AMTS).9
Table 1
shows the application of the ambulatory case mix methodology, with the figures in parentheses indicating the number of patients in each subgroup. Table 2 shows the demographic characteristics in each MAC; the breakdown of the workload within each MAC in terms of the number of all visits and problems encountered; the relative workload within each consultation; and gender and age group differences.
|
|
The pattern of distribution of patients and the related workload across all MACs allowed the construction of general practice-specific care categories (GP-SCCs). These categories are defined as: GP-SCC 1, acute/self-limiting problems and preventive care (MAC 1, 5 and 8); GP-SCC 2, primarily chronic health problems (MAC 2, 3, 6, 7 and 9); GP-SCC 3, psychosocial problems in conjunction with up to two other categories (MAC 4, 10, 11 and 12); and GP-SCC 4, a combination of four or more categories (MAC 13). Table 3
summarizes patient and workload characteristics for each GP-SCC.
|
During the 2-year study period, doctors conducted
44 000 consultations. The average patient consulted 10.67 times (CI 9.62–11.77), and discussed an average of 13.19 problems (CI 11.74–14.63). The mean number of problems encountered per consultation was 1.23 (CI 1.20–1.26). To determine the workload implications for GPs, three parameters were compared: proportion of patients; proportion of visits; and proportion of problems encountered in each GP-SCC.
Patients presenting with self-limiting disease (GP-SCC 1) comprised 31.1% (CI 29.1–35.1) of patients, accounting for 25.6% (CI 24.0–27.3) of visits and 21.9% (CI 20.5–23.3) of all problems encountered. Patients in GP-SCC 2, i.e. those with limited chronic disease, comprised 16.7% (CI 10.6–19.6) of patients, accounting for 10.6% (CI 9.5–11.9) of visits and 9.9% (CI 8.9–11.0) of all problems encountered. Patients with psychological and some limited disease (GP-SCC 3) comprised 7.1% (CI 4.4–11.2) of patients, accounting for 7.8% (CI 6.8–8.9) of visits and 7.7% (CI 6.8–8.7) of all problems encountered. Finally, patients from GP-SCC 4, i.e. those with multisystem problems, comprised 42.0% (CI 35.8–48.2) of all patients, accounting for 56.0% (CI 54.2–57.8) of all visits and 60.5% (CI 58.8–62.2) of all problems encountered.
Comparing the workload in each GP-SCC against the mean workload of the study population reveals that patients with acute problems (GP-SCC 1) had on average 2.87 (27.8%) fewer visits and presented with 4.98 (37.8%) fewer problems, and patients with limited chronic disease (GP-SCC 2) had 3.16 (29.6%) fewer visits and presented with 4.54 (34.4%) fewer problems. Patients with psychological and other limited disease (GP-SCC 3) created the average workload, whereas patients with multisystem problems (GP-SCC 4) had 3.56 (33.4%) more visits and presented with 5.80 (44.0%) more problems.
The workload per consultation expressed as the ratio of all problems encountered during all visits shows a similar difference between the four groups, with GP-SCC 1 having a mean number of problems per consultation of 1.05 (CI 1.02–1.07; 11.7% lower), GP-SCC 2 of 1.15 (CI 1.09–1.24; 2.5% lower), GP-SCC 3 of 1.21 (CI 1.13–1.29; average) and GP-SCC 4 of 1.33 (CI 1.28– 1.39; 11.7% higher).
Results indicate that males consult more frequently with acute/self-limiting problems (GP-SCC 1), whilst females tend to present more frequently with psychological (GP-SCC 3) and multisystem problems (GP-SCC 4).
The number of visits and problems steadily increase across the GP-SCC groups and with age. Those over 65 years old visited most frequently and presented with the largest number of health complaints.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionsReferences |
|---|
Statistics by the Health Insurance Commission10 for the year ending June 1997 estimate that, on average, Australians consulted a GP 5.48 times. Not everyone, however, seeks medical care during the year, which increases the figure to 6.32 visits for those who actually attended a GP.
Representativeness of morbidity in the sample
The illness distribution and frequency of problems managed are similar to those of the AMTS.9 Differences in the acute illness groups are explained by difficulties in applying labels to undifferentiated problems, e.g. ‘sore throat’ may mean tonsillitis, viral illness or URTI, whereas ‘otitis media’ is a distinct diagnosis to which a label can be applied easily and consistently between providers. The slightly higher proportion of the elderly in the sample explains the higher rate of chronic disease, whereas the incidence of depression is comparable across both study populations (Table 4).
|
General practice-specific care categories
MACs were designed to measure the utilization of health care services by different patient groups in a health insurance scheme in the USA. Based on the model, researchers were able to predict the future use of resources for each category. These groupings are too small to be useful for the general practice setting. The GP-SCC model categorizes patients according to their experienced morbidity. The model is a useful tool for individual practices to examine their practice population' health and the associated workload. Retrospective chart audits have the advantage of identifying the health needs for which care was sought, though poor recording potentially may underestimate the true need.
Workload
Quantifying workload is difficult, and several factors may impact on the measure, including accessibility of the provider, doctor–patient ratio, socio-economic environment and age structure of the population. In this study, we looked at three parameters that impact on the workload of a GP: frequency of attendance; number of problems encountered; and the resulting relative workload per consultation expressed as the ratio of problems encountered in each visit. From this study, it appears that the latter best reflects the demand on a GP.
The perceived need to attend a doctor is not determined solely by the patient's actual diseases. ‘Feeling ill’ is a strong motivator to seeking medical care, regardless of the presence or absence of disease.11 In fact, research by Gulbrandsen and colleagues12 demonstrated that one-third of all consultations are motivated primarily by a psychosocial health problem.
In overall terms, attendance frequencies and health problems encountered increased with age, as one would expect. The difference in the underlying illness and disease burden across the age spectrum is obvious, the young attending primarily with acute self-limiting disease and for preventive health care, the older with multiple, mostly chronic, health problems.
Gender differences in health concern were evident, with females exhibiting a greater concern when having multisystem disease and discussing more psychosocial health issues. This study clearly confirms the finding that psychosocial health problems increase the demand for health care, hence dramatically increasing workload. Subanalysis showed that 67 (27.3%) patients had a psychosocial health problem presenting on average 15.55 (CI 12.60–18.50) times [8.87 (CI 8.84–9.71) visits for those without; P < 0.0001] with 20.40 (CI 16.49–24.31) problems [10.47 (CI 9.37–11.57) problems for those without; P < 0.0001]. This pattern remains independent of the level of co-morbidity: 46.8% (CI 37.3–56.3) of those in GP-SCC 4 had a psychosocial health problem, visiting on average 17.44 (CI 13.83–21.05) times and presenting with 23.30 (CI 18.59–28.01) problems. Those without psychosocial health problems had 11.44 (CI 9.50–13.38) visits (P = 0.007) and presented with 15.24 (CI 10.53–19.95) problems (P = 0.01).
It is interesting to note that there is little difference in workload per consultation across the age spectrum within MACs and GP-SCCs, which indicates that the illness experience as such determines the need to seek care.11
Can the findings be generalized?
This study explored the usefulness of the ambulatory case mix concept to describe the workload of Australian general practice. The data presented are representative of a medium sized practice population and demonstrate the feasibility of the methodology. However, to allow generalizability, the methodology needs to be applied to a larger number of practices in different community settings.
Data reliability may be limited by using a retrospective chart audit process;13,14 however, it has been shown that medical records adequately reflect what actually happened in the consultation.15,16
The strength of this study is its approach of following individual patients over time without their behaviour being influenced by the study, an approach not used previously in other studies examining general practice workload patterns.9,17
The impact of co-morbidity
Little is known about the ‘true’ workload experienced by GPs. This study adds to our understanding of the impact that co-morbidities have on the demands within each consultation. The GP-SCCs are a reliable proxy indicator of workload. Its usefulness in terms of planning for workforce distribution and practice infrastructure in different communities should be examined further.
Australian GPs already spend a large amount of their time on patients with multisystem disease. The new initiatives, aimed to improve the care of this group of patients, further increase the pressures on the resource ‘provider’. As stated above, new initiatives are needed18 to improve quality of care and quality of life, but in light of these findings appear achievable and sustainable only by increasing the number of providers.
Future work will be carried out to relate these findings to the actual cost of care for each of the GP-SCCs.
|
Conclusions |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionsReferences |
|---|
The general practice-specific care category model, built on the ambulatory case mix concept, is a useful tool to analyse the morbidity of practice populations, and has a good predictive value in terms of a practice's workload.
|
Acknowledgments |
|---|
My special thanks go to Professor Barbara Starfield who kindly appraised the development of the GP-SCC methodology, and Kristie Chamberlain for her help with the preparation of the manuscript.
|
References |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionsReferences |
|---|
1 National Prescribing Service Ltd (NPS). Level 1/31 Buckingham St, Surry Hills 2010, Australia. www.nps.org.au
2 Better Health Care for Australians: New Initiatives in Primary Health Care. Fact Sheet 2. Budget 1999–2000. Commonwealth Department of Health and Aged Care. www.health.gov.au/pubs/budget99/fact/hfact2.htmEnhanced Primary Care and the Federal Government's New MBS Items. Fact Sheet http://partners.health.gov.au/hfs/hsdd/primcare/enhancpr/mbsitems.htmA Practical Glance at the New MBS Items. http://partners.health.gov.au/hfs/hsdd/primcare/enhancpr/mbsglanc.htm
3 Marcus D. Coordinating Care in an Uncoordinated Health System: The Development and Implementation of Coordinated Care Trials in Australia. Current Issues Brief 11 1998–99. www.aph.gov.au/library/pubs/cib/1998-99/99cib11.htm
4 Weiner JP, Starfield BH, Steinwachs DM, Mumford LM. Development and application of a population-oriented measure of ambulatory care case-mix. Med Care 1991; 29:452–472.[Web of Science][Medline]
5 Starfield B, Weiner J, Mumford L, Steinwachs D. Ambulatory care groups: a categorization of diagnoses for research and management. HSR Health Serv Res 1991; 26:53–74.
6 SPSS 10.0TM, Chicago (IL): SPSS Inc., 1999.
7 Sturmberg JP. Masters Thesis. Melbourne: Monash University, 1998.
8 Sturmberg JP, Schattner P. Personal doctoring—its impact on continuity of care as measured by the comprehensiveness of care score. Aust Fam Physician 2001; 30:513–518.[Medline]
9 Bridges-Webb C, Britt H, Miles DA, Neary S, Charles J, Traynot V. Morbidity and treatment in general practice in Australia 1990–1991. Med J Aust 1992; 157 (Suppl):S1–S57.
10 Health Insurance Commission. Annual Report 1996–97. Canberra: Health Insurance Commission, 1997.
11 Ridsdale L, Mandalia S and the General Practice Fatigue Group. Tiredness as a ticket of entry. The role of patients' beliefs and psychological symptoms in explaining frequent attendance. Scand J Prim Health Care 1999; 17:72–74.[Web of Science][Medline]
12
Gulbrandsen P, Fugelli P, Hjortdahl P. Psychosocial problems presented by patients with somatic reasons for encounter: tip of the iceberg? Fam Pract 1998; 15:1–8.
13 Gordon J, Kearney M, Watson P. Medical records in general practice. Aust Med J 1992; 156:701–709.
14 Rethans JJ, Martin E, Metsemakers J. To what extent do clinical notes by general practitioners reflect actual medical performance? A study using simulated patients. Br J Gen Pract 1994; 44:153–156.[Web of Science][Medline]
15
Zuckerman AE, Starfield B, Hochreiter C, Kovaszrnay B. Validating the content of pediatric outpatient medical records by means of tape-recording doctor–patient encounters. Pediatrics 1975; 56:407–411.
16 Borgiel AEM, Williams JI, Anderson GM et al. Assessing the quality of care in family physicians' practices. Can Fam Physician 1985; 31:853–862.
17 Britt H, Sayer GP, Miller GC, Charles J et al. General Practice Activity in Australia 1998–99. AIHW Cat. No. GEP 2. Canberra: Australian Institute of Health and Welfare (General Practice Series no. 2), 1999.
18 Bain J. All good doctors should . . . (Editorial). J R Coll Gen Pract 1986; 36:249–251.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  E. M Zantinge, P. F. Verhaak, and J. M Bensing The workload of GPs: patients with psychological and somatic problems compared Fam. Pract., June 1, 2005; 22(3): 293 - 297. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||