Abstract
Objective
To determine whether the priority given to patients referred for cardiac surgery is associated with socioeconomic status.
Design
Retrospective study with multivariate logistic regression analysis of the association between deprivation and classification of urgency with allowance for age, sex, and type of operation. Multivariate linear regression analysis was used to determine association between deprivation and waiting time within each category of urgency, with allowance for age, sex, and type of operation.
Setting
NHS waiting lists in Scotland.
Participants
26 642 patients waiting for cardiac surgery, 1 January 1986 to 31 December 1997.
Main outcome measures
Deprivation as measured by Carstairs deprivation category. Time spent on NHS waiting list.
Results
Patients who were most deprived tended to be younger and were more likely to be female. Patients in deprivation categories 6 and 7 (most deprived) waited about three weeks longer for surgery than those in category 1 (mean difference 24 days, 95% confidence interval 15 to 32). Deprived patients had an odds ratio of 0.5 (0.46 to 0.61) for having their operations classified as urgent compared with the least deprived, after allowance for age, sex, and type of operation. When urgent and routine cases were considered separately, there was no significant difference in waiting times between the most and least deprived categories.
Conclusions
Socioeconomically deprived patients are thought to be more likely to develop coronary heart disease but are less likely to be investigated and offered surgery once it has developed. Such patients may be further disadvantaged by having to wait longer for surgery because of being given lower priority.
Introduction
Socioeconomic deprivation is associated with both prevalence of and mortality from coronary heart disease.1–3 Social class differences in mortality from coronary heart disease have widened over the past three decades.4 Despite being at greater risk of developing coronary heart disease and dying from it, patients in lower socioeconomic groups are less likely to be investigated once the disease develops5–10 and are less likely to be referred for cardiac surgery thereafter.5–12
We studied whether socioeconomic inequalities also exist in the priority given to patients on the waiting list for cardiac surgery.
Methods
In Scotland information is routinely collected on every patient who is added to the waiting list for cardiac surgery by using the Scottish Morbidity Record 20 (SMR20) system. The Information and Statistics Division of the Common Services Agency in Edinburgh collates these data. The division provided SMR20 data on all patients on the cardiac surgery waiting list from 1 January 1986 to 31 December 1997. The information included age, sex, urgency, type of operation, dates of entry on to and exit from the waiting list, date of surgery, and postcode. The postcodes were used to derive Carstairs socioeconomic deprivation categories.13 These range from 1 to 7 and are based on 1991 census data on car ownership, unemployment, overcrowding, and social class within postcodes. Category 1 denotes the least deprived areas and 7 the most deprived.
Multivariate logistic regression analysis was used to determine whether the deprivation category was associated with surgery being classified as urgent, after allowance for age, sex, and type of operation. Multivariate linear regression analysis was used to determine whether the deprivation category was associated with waiting time within each category of urgency, after allowance for age, sex, and type of operation.
Results
In total 26 642 patients were placed on the waiting list for cardiac surgery over the period studied. Socioeconomic deprivation was associated with a greater likelihood that the patient was female (P<0.0001) and under 65 years of age (P<0.0001) (table 1). The mean waiting time for surgery increased across the deprivation categories, with patients in categories 6 and 7 (most deprived) waiting, on average, three weeks longer than those in category 1 (P<0.0001) (table 2). There was a significant association between socioeconomic deprivation and classification of urgency. Only 22% of patients in categories 6 and 7 were classified as urgent compared with 36% of those in category 1 (P<0.0001; table 1). When account was taken of age, sex, and type of operation patients in categories 6 and 7 had an odds ratio of 0.5 for being classified as urgent (table 3). The mean waiting times for routine and urgent cases were 196 days and 67 days, respectively.
Table 1.
Case mix of patients on waiting list for cardiac surgery by socioeconomic deprivation category (1=least deprived to 7=most deprived). Figures are percentages of patients*
| Detail | No of patients | 1 (n=1541) | 2 (n=3409) | 3 (n=5451) | 4 (n=6515) | 5 (n=4191) | 6 (n=3355) | 7 (n=2180) |
|---|---|---|---|---|---|---|---|---|
| Age (years): | ||||||||
| <55 | 7 386 | 22.5 | 23.9 | 26.4 | 27.3 | 30.1 | 28.5 | 36.2 |
| 55-64 | 10 885 | 38.7 | 40.0 | 39.6 | 41.7 | 41.7 | 41.8 | 41.4 |
| >64 | 8 371 | 38.7 | 36.1 | 34.0 | 31.0 | 28.3 | 29.7 | 22.4 |
| Sex: | ||||||||
| Male | 18 706 | 73.0 | 73.1 | 73.1 | 69.1 | 69.8 | 66.3 | 66.9 |
| Female | 7 936 | 27.0 | 26.9 | 26.9 | 30.9 | 30.2 | 33.7 | 33.1 |
| Type of operation: | ||||||||
| CABG only | 20 213 | 77.9 | 75.7 | 76.9 | 75.0 | 75.6 | 74.2 | 77.6 |
| CABG+valve(s) | 742 | 2.5 | 2.8 | 2.0 | 2.6 | 2.7 | 4.5 | 3.0 |
| Single valve | 5 141 | 18.4 | 19.7 | 19.6 | 20.1 | 19.4 | 18.7 | 17.1 |
| Multiple valves | 546 | 1.2 | 1.8 | 1.5 | 2.2 | 2.3 | 2.6 | 2.3 |
| Urgency: | ||||||||
| Routine | 17 790 | 63.7 | 62.2 | 61.0 | 64.7 | 67.5 | 78.3 | 77.7 |
| Urgent | 8 852 | 36.3 | 37.8 | 39.0 | 35.3 | 32.5 | 21.7 | 22.3 |
CABG=coronary artery bypass graft
Percentages relate to breakdown of each deprivation category by age, sex, type of operation, and urgency.
Table 2.
Multivariate linear regression analysis of factors associated with waiting times in days for cardiac surgery for all patients
| Detail | No of patients | Mean (SD) waiting time (days) | Difference in mean waiting time* (95% CI) |
|---|---|---|---|
| Age: | |||
| <55† | 6 906 | 143.7 (126.0) | 0.0 |
| 55-64 | 10 060 | 158.2 (135.0) | 14.6 (10.7 to 18.8) |
| >64 | 7 597 | 150.8 (135.6) | 10.8 (6.4 to 15.1) |
| Sex: | |||
| Male† | 17 251 | 152.6 (133.4) | 0.0 |
| Female | 7 312 | 150.0 (131.4) | 2.8 (−1.0 to 6.7) |
| Operation type: | |||
| CABG† | 18 613 | 157.4 (133.8) | 0.0 |
| CABG+valve(s) | 660 | 135.5 (133.2) | −25.1 (−14.8 to −35.4) |
| Single valve | 4 790 | 134.8 (131.1) | −23.6 (−19.2 to −28.0) |
| Multiple valves | 500 | 130.7 (120.8) | −28.4 (−16.6 to −40.3) |
| Carstairs deprivation category: | |||
| 1† | 1 418 | 139.7 (130.9) | 0.0 |
| 2 | 3 133 | 144.9 (133.2) | 5.9 (−2.4 to 14.2) |
| 3 | 5 040 | 149.4 (135.1) | 10.5 (2.7 to 18.3) |
| 4 | 6 028 | 152.8 (133.3) | 14.2 (6.59 to 21.9) |
| 5 | 3 852 | 150.4 (134.7) | 12.0 (4.0 to 20.1) |
| 6 | 3 100 | 161.8 (129.0) | 23.5 (15.1 to 31.8) |
| 7 | 1 992 | 161.9 (127.1) | 23.5 (14.5 to 32.5) |
Compared with reference categories adjusted for the other factors in the model.
Reference categories.
Table 3.
Univariate and multivariate logistic regression analyses of factors associated with urgent rather than routine priority on cardiac surgery waiting list. Figures are odds ratios (95% confidence intervals)
| Detail | No of patients | Univariate analysis | Multivariate analysis |
|---|---|---|---|
| Age: | |||
| <55* | 7 386 | 1.00 | 1.00 |
| 55-64 | 10 885 | 1.00 (0.94 to 1.06) | 0.99 (0.93 to 1.05) |
| >64 | 8 371 | 1.44 (1.35 to 1.54) | 1.35 (1.26 to 1.45) |
| Sex: | |||
| Male* | 18 706 | 1.0 | 1.0 |
| Female | 7 936 | 1.05 (1.00 to 1.11) | 0.93 (0.87 to 0.98) |
| Operation type: | |||
| CABG* | 20 213 | 1.0 | 1.0 |
| CABG+valve(s) | 772 | 1.60 (1.50 to 1.70) | 1.57 (1.47 to 1.68) |
| Single valve | 5 141 | 1.13 (0.94 to 1.35) | 1.22 (1.02 to 1.47) |
| Multiple valves | 546 | 0.68 (0.58 to 0.81) | 0.69 (0.58 to 0.83) |
| Carstairs deprivation category: | |||
| 1* | 1 541 | 1.00 | 1.0 |
| 2 | 3 409 | 1.07 (0.94 to 1.21) | 1.07 (0.94 to 1.21) |
| 3 | 5 451 | 1.12 (1.00 to 1.26) | 1.13 (1.01 to 1.27) |
| 4 | 6 515 | 0.96 (0.85 to 1.07) | 0.97 (0.87 to 1.09) |
| 5 | 4 191 | 0.84 (0.74 to 0.95) | 0.86 (0.76 to 0.98) |
| 6 | 3 355 | 0.49 (0.43 to 0.55) | 0.50 (0.44 to 0.57) |
| 7 | 2 180 | 0.50 (0.43 to 0.58) | 0.53 (0.46 to 0.61) |
CABG = coronary artery bypass graft.
Reference categories.
When routine and urgent cases were considered separately the association between waiting time and deprivation category was an inverted U shape rather than linear (table 4). Waiting times were lowest in the most and least deprived categories of patients and highest in the middle groups, with no significant difference between categories 7 and 1.
Table 4.
Multivariate linear regression analyses of factors associated with mean* (SD) waiting time in days for cardiac surgery for routine and urgent cases
| Detail | Routine cases
|
Urgent cases
|
|||||
|---|---|---|---|---|---|---|---|
| No of patients | Mean (SD) | Difference in mean waiting time† (95% CI) | No of patients | Mean (SD) | Difference in mean waiting time† (95% CI) | ||
| Age (years): | |||||||
| <55‡ | 4 725 | 180.7 (127.0) | 0.0 | 2181 | 63.6 (76.9) | 0.0 | |
| 55-64 | 6 896 | 198.9 (130.1) | 18.2 (13.4 to 23.0) | 3164 | 69.6 (97.9) | 6.0 (1.1 to 10.9) | |
| >64 | 4 538 | 206.6 (133.0) | 28.5 (23.1 to 33.9) | 3059 | 67.9 (89.2) | 4.0 (−1.0 to 9.0) | |
| Sex: | |||||||
| Male‡ | 11 386 | 196.5 (130.9) | 0.0 | 5865 | 67.3 (89.8) | 0.0 | |
| Female | 4 773 | 193.8 (129.2) | 2.1 (−2.6 to 6.8) | 2539 | 67.7 (89.3) | −0.9 (−5.3 to 3.5) | |
| Operation type: | |||||||
| CABG‡ | 12 596 | 201.0 (130.1) | 0.0 | 6017 | 66.1 (87.1) | 0.0 | |
| CABG+valve(s) | 496 | 158.9 (115.3) | −46.9 (−58.6 to −35.2) | 164 | 64.7 (68.6) | −1.2 (−15.2 to 12.8) | |
| Single valve | 2 741 | 181.7 (132.3) | −21.8 (−27.4 to −16.2) | 2049 | 72.0 (99.4) | 5.6 (1.1 to 10.5) | |
| Multiple valves | 326 | 167.3 (126.2) | −33.7 (−48.1 to −19.3) | 174 | 62.1 (70.3) | −3.8 (−17.4 to 10.2) | |
| Carstairs deprivation category: | |||||||
| 1‡ | 886 | 192.7 (132.7) | 0.0 | 532 | 51.6 (62.4) | 0.0 | |
| 2 | 1 902 | 197.5 (129.3) | 6.0 (−4.3 to 16.3) | 1231 | 63.7 (91.9) | 12.0 (2.9 to 21.1) | |
| 3 | 3 029 | 201.2 (136.2) | 10.0 (0.3 to 19.7) | 2011 | 71.4 (87.6) | 19.6 (11.0 to 28.2) | |
| 4 | 3 837 | 199.5 (133.3) | 9.6 (0.1 to 19.1) | 2191 | 70.9 (85.1) | 19.3 (10.8 to 27.8) | |
| 5 | 2 559 | 191.2 (126.9) | 2.0 (−7.9 to 11.9) | 1293 | 69.6 (111.2) | 18.0 (8.9 to 27.1) | |
| 6 | 2 419 | 189.0 (126.2) | −0.3 (−9.7 to 10.3) | 681 | 64.8 (84.4) | 14.0 (3.8 to 24.1) | |
| 7 | 1 527 | 193.0 (122.9) | 4.4 (−6.4 to 15.2) | 465 | 59.6 (77.1) | 8.6 (−2.6 to 19.8) | |
Unadjusted for other factors in model.
Compared with reference categories adjusted for other factors in model.
Reference categories.
Discussion
Mortality and morbidity from coronary heart disease show a social class gradient, with more deprived groups experiencing a greater burden of disease.1–3 In men the mortality from coronary heart disease is 40% higher in manual than non-manual workers.1 Wives of manual and non-manual workers experience a twofold difference.1 In contrast with population mortality,1–3 case fatality does not vary significantly by socioeconomic group.3
Although social inequalities in coronary heart disease have been found in most countries, they vary in magnitude. The United Kingdom has a much higher social class gradient than some other countries, such as Sweden.14 The overall mortality from coronary heart disease has declined over the past three decades. The decline, however, has been greater in the most affluent groups.4 As a result, the social class gradient in such mortality has increased.
Despite being more likely to develop coronary heart disease and die from it, patients in lower socioeconomic groups are less likely to be investigated with coronary angiography once the disease develops5–10 and are also less likely to be referred for coronary artery bypass grafting.5–12
Our results suggest that after referral for cardiac surgery, more deprived patients may be disadvantaged further in that they are required to wait significantly longer for their operations. This results primarily from the fact that the most affluent patients were significantly more likely to have their operations classified as urgent compared with the least affluent patients. Overall, the most deprived patients were required to wait three weeks longer for surgery. An additional waiting time of this magnitude may not be clinically important for routine cases. Deprived patients, however, had only half the odds of being classified as urgent cases. Urgent cases were, on average, operated on 129 daysearlier than routine cases. An excess delay of this magnitude due to differences in classification of urgency may be associated with more frequent adverse events on the waiting list.
Study limitations
The SMR20 dataset does not collect information on the severity of cardiac disease and the presence of comorbidity. Obviously both of these need to be considered in determining whether waiting times accurately reflect clinical need and risk. Lack of these data constitute a limitation of this study, and therefore care should be taken in drawing conclusions. As deprived patients with coronary heart disease are less likely to be investigated and referred for surgery at the outset,5–12 however, it is likely that those deprived patients who are added to the waiting list have more severe cardiac disease than their more affluent counterparts. As a result, intuition would suggest that prioritisation by clinical need should favour socioecononically deprived patients. Therefore, it is likely that this study underestimates the extent to which more affluent patients are favoured. In addition to severity of cardiac disease and comorbidity, decisions on priority may take account of non-clinical factors such as employment status and dependants. Data on these factors were unavailable for analysis, and the extent to which they do and should contribute to priority setting is subject to debate.
What is already known on this topic
Socioeconomic deprivation is associated with a greater likelihood of developing coronary heart disease
Although deaths from the disease have declined over the past three decades, this decline has been greatest in the most affluent groups, and as a result the social class gradient in mortality has increased
Lower socioeconomic groups are less likely to be investigated once coronary heart disease develops and are less likely to be referred for cardiac surgery
What this paper adds
On average, the most deprived patients waited about three weeks longer for surgery than the most affluent
Deprived patients had an odds ratio of 0.5 for having their operations classified as urgent, after allowance for age, sex, and type of operation
When urgent and routine cases were considered separately there was no significant difference in waiting times between the most and least deprived categories
In addition to their greater burden of disease, worse prognosis, and poorer access to investigation and surgery, socioeconomically deprived patients may be further disadvantaged by having to wait longer for surgery because of being given lower priority
Once classified into urgent and routine cases the differences in waiting time between the most and least deprived categories were no longer significant. Both groups, however, waited significantly less time than those in the middle categories. This may reflect a combination of factors. Possibly, compared with the most and least deprived groups, those in the middle may be less clinically needy and less vocal, respectively. Data to substantiate or refute this hypothesis, however, were not available from this study.
These results add to the growing evidence of socioecononic inequalities in health care. Previous studies suggest that, despite being more likely to develop coronary heart disease, socioeconomically deprived patients are less likely to be investigated and treated. Even after treatment is offered, deprived patients may be further disadvantaged by being required to wait longer for surgery.
Acknowledgments
We are grateful to the Information and Statistics Division of the Common Services Agency for providing the data on which the analyses were undertaken.
Footnotes
Funding: Chief Scientist's Office of the Scottish Executive's Department of Health.
Competing interests: None declared.
References
- 1.Kaguskar S, Bradshaw H, Rayner M. Coronary heart disease statistics. London: British Heart Foundation; 1997. [Google Scholar]
- 2.Drever F, Whitehead M. Health inequalities: decennial supplement. London: Stationery Office; 1997. [Google Scholar]
- 3.Office of Population Censuses and Surveys. 1991 census. London: HMSO; 1995. [Google Scholar]
- 4.Drever F, Whitehead M, Roden M. Current patterns and trends in male mortality by social class based on occupation. Pop Trends. 1996;86:15–20. [PubMed] [Google Scholar]
- 5.Manson-Siddle CJ, Robinson MB. Super Profile analysis of socioeconomic variations in coronary investigation and revascularisation rates. J Epidemiol Community Health. 1998;52:507–512. doi: 10.1136/jech.52.8.507. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Findlay IN, Dargie HJ, Dyke T, Archibald M. Who gets coronary angiography in Scotland? Br Heart J. 1990;64:43–44. [Google Scholar]
- 7.Findlay IN, Dargie JH, Dyke T. Coronary angiography in Glasgow: relation to coronary heart disease and social class. Br Heart J. 1991;66:A70. [Google Scholar]
- 8.Gittelsohn AM, Halalpern J, Sanchez RL. Income, race, and surgery in Maryland. Am J Public Health. 1991;81:1435–1441. doi: 10.2105/ajph.81.11.1435. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Payne N, Saul C. Variations in use of cardiology services in a health authority: comparison of coronary artery revascularisation rates with prevalence of angina and coronary mortality. BMJ. 1997;314:257–261. doi: 10.1136/bmj.314.7076.257. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Keskimaki I, Koskinen S, Salinto M, Aro S. Socioeconomic and gender inequities in access to coronary artery bypass grafting in Finland. Eur J Pub Health. 1997;7:392–397. [Google Scholar]
- 11.Ben-Shlomo Y, Chaturvedi N. Assessing equity in access to health care provision in the UK: does where you live affect your chances of getting a coronary artery bypass graft? J Epidemiol Community Health. 1995;49:200–204. doi: 10.1136/jech.49.2.200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.MacLeod MCM, Finlayson AR, Pell JP, Findlay IN. Geographical, demographic and socioeconomic variations in the investigation and management of coronary heart disease in Scotland. Heart. 1999;81:252–256. doi: 10.1136/hrt.81.3.252. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Carstairs V, Morris R. Deprivation and health in Scotland. Aberdeen: Aberdeen University Press; 1991. [Google Scholar]
- 14.Vagero D, Lundgren O. Health inequalities in Britain and Sweden. Lancet. 1989;299:35–36. doi: 10.1016/s0140-6736(89)90266-3. [DOI] [PubMed] [Google Scholar]