Family Practice Advance Access originally published online on April 3, 2006
Family Practice 2006 23(3):295-302; doi:10.1093/fampra/cml003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
A cross-sectional study of secondary cardiac care in general practice: impact of personal and practice characteristics
a Department of General Practice, National University of Ireland Galway
b Department of Psychology, National University of Ireland Galway
c Department of Public Health, North Western Health Board, University College Dublin Ireland
d Department of Public Health and Epidemiology, University College Dublin Ireland
Correspondence to Professor Andrew W Murphy, Department of General Practice, Clinical Sciences Institute, National Univesity of Ireland, Galway, Ireland; Email: andrew.murphy{at}nuigalway.ie
Received 29 April 2005; Accepted 8 March 2006.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodologyResultsDiscussionReferences |
|---|
Objective. To determine the impact of patient (age, gender, type of and time since diagnosis) and practice (rurality, number of partners, availability of practice nurse) characteristics on secondary cardiac care in general practice in a country without universal registration.
Methods. Medical and demographic data were gathered from the medical charts of 1611 eligible patients from 35 randomly selected practices. Eligible patients were aged under 80 years with a recorded history of acute myocardial infarction, percutaneous trans coronary arteriogram or angina. Self-report data about diet, exercise, smoking and alcohol consumption were provided from postal questionnaire (1084 patients responded; 69% response rate).
Results. Having an angina only diagnosis significantly decreased the likelihood of patients being prescribed aspirin (OR = 0.53; 95% CI = 0.40–0.69), lipid-lowering medication (OR = 0.55; 95% CI = 0.43–0.69) or ACE inhibitors (OR = 0.62; 95% CI = 0.48–0.81). Younger patients (OR = 1.05; 95% CI = 1.04–1.06) were also more likely to be prescribed lipid-lowering medication. Cholesterol was predicted by gender only, with females having significantly higher cholesterol (B = –0.41; 95% CI = –0.54 to –0.27). The number of missed opportunities for secondary cardiac care was greater among patients with angina only (B = 0.39; 95% CI = 0.19–0.60). The amount of variance explained by practice and patient variables overall for each of the measures was small, ranging from 2 to 6%.
Conclusions. Practice size or location appears to have little impact on secondary cardiac care. The most consistent significant personal characteristic finding was that patients with a diagnosis of angina only were significantly less likely to receive aspirin, statins or ACE inhibitors and more likely to have more missed opportunities for secondary cardiac care.
Keywords. Angina, arteriogram, lipid lowering, statin, secondary prevention, coronary heart disease.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodologyResultsDiscussionReferences |
|---|
The secondary prevention of cardiac disease within general practice is both effective and feasible.1 The provision of such care to patients with established heart disease has often been shown to be suboptimal.2,3 The characteristics of individual health systems, providers and patients may all potentially impact on the provision of such care. Both the management of, and research about, chronic disease within the community is enhanced by universal patient registration.4 Many of the frequently cited studies5–8 concerning secondary cardiac care originate from the UK which utilizes universal patient registration. However, not all health systems utilize patient registration and the generalizabilty of research between such different systems needs to be considered. The relationships between practice characteristics of size, availability of practice nursing and rurality and secondary cardiac care have received little attention. The impact of patient characteristics have been largely studied through large routinely collected computerized medical databases, with a particular focus on statin prescribing.9–13 These studies have generally suggested that older and female patients with a diagnosis of angina only are less likely to receive optimal treatment.
We therefore considered it opportune to determine the impact of patient (age, gender, type of and time since diagnosis, socioeconomic status) and practice (rurality, number of partners, availability of practice nurse) characteristics on levels of secondary cardiac care from a stratified, random sample of general practices in a country without universal registration.
|
Methodology |
|---|
|
TopAbstractIntroductionMethodologyResultsDiscussionReferences |
|---|
The study was conducted in the North Western and Western Health Boards in the western seaboard region of Ireland. Between them the two boards cover an area of 13 463 sq km and serve a population of 573 250.
Free primary care and medications are available to only 29% of the population of the Republic of Ireland; they are described as General Medical Services (GMSs) eligible. The remainder (referred to as ‘private patients’), whose income is above a certain level, are responsible for their own primary health care costs. GMS eligible patients therefore represent the least affluent sector of the community. Whilst a ‘panel list’ is available for all GMS eligible patients, practices do not have complete registers of regularly attending private patients.
Practice recruitment
Lists of all GPs for the two boards were obtained and practices stratified according to whether they were single-handed or partnership, and rural or urban. Using random numbers, practices were selected in May 2000 from the list until the required number of practices had been achieved, ensuring that the stratifying criteria were satisfied.
Patient eligibility, recruitment and exclusion
GPs were asked to generate a list of all their patients with established heart disease (Supplementary Table 1 available online at www.fampr.oxfordjournals.org). These were defined as patients with a history of myocardial infarction (MI); angina; or revascularization by angioplasty (PTCA) or coronary artery bypass grafting (CABG). Patients were excluded from the study for the following reasons: over the age of 80; terminally ill; housebound with serious co-morbidity; known to have moved away or at the discretion of the GP.
The Second Joint Task Force of European and other Societies on Coronary Prevention recommendations14 had been adopted in Ireland at the time of study conduct and these were applied to the results (Supplementary Table 2 available online at www.fampr.oxfordjournals.org). Once data had been collected from patient charts by one of two trained research nurses, questionnaires were posted in January 2001 to patients. Questionnaires obtained self-report data on perceived general health, smoking, diet, alcohol consumption, exercise, adherence to medication and perceptions of illness and treatment. Results regarding patient perceptions of illness and treatment have been published elsewhere.15
Exercise behaviour was assessed using the Godin Leisure Time Exercise Questionnaire.16 A score of 15 or higher on the questionnaire indicated that the patient was exercising at the recommended level. Body mass index (BMI) was calculated from patient self-reports of weight and height. Respondents' dietary habits were assessed using the Dietary Instrument for Nutrition Education (DINE)17 which provided scores for fat and fibre intake.
As performed previously by Campbell,5 cumulative ratings (one point each) for medical management and lifestyle were calculated for each patient according to the number of missed opportunities for secondary prevention based on the European guidelines.14 Missed medical management opportunities referred to suboptimal aspirin treatment, blood pressure above recommended level or cholesterol above recommended level. Missed lifestyle referred to suboptimal exercise, current smoking or high fat diet.
Main outcome measurements
The main outcome measurements were categorized as continuous (blood pressure, cholesterol, exercise score, dietary fat and number of missed opportunities) or dichotomous (prescription of aspirin, lipid-lowering agent and ACE inhibitor).
Sample size
Utilizing estimated proportions of main outcome measures from the Grampian10 study (e.g. 17% of patients in target range for cholesterol), and a questionnaire response rate of 65%, we calculated that a total sample of 1548 patients would provide an estimated proportion of 16.99% (95% CI 14.72–19.45) which appeared reasonable.
Statistical analysis
Data were stored and analysed using SPSS statistical software, version 11.
Univariate analyses were conducted first to examine the independent relationships between independent variables and secondary prevention outcome variables. Multivariate analyses were then carried out to examine the relative contribution of independent variables in predicting secondary prevention outcome variables. Hierarchical regression analyses were carried out, in which practice characteristics were inserted as the first block and patient characteristics were inserted as the second block of predictors, to compare their relative contributions with predicting secondary prevention outcomes.
A series of hierarchical multiple regression analyses using a random effects model were conducted to assess the impact of explanatory variables divided into two levels, namely practice (location, partner versus single-handed, practice nurse and GMS size) and patient characteristics (age, gender, GMS eligibility, time since diagnosis and angina), for the different outcomes (i.e. systolic and diastolic BP, level of cholesterol, exercise, dietary fat and number of missed opportunities).
A series of hierarchical logistic regression analyses were also run to examine impact of the same practice variables (entered in Block 1) and patient variables (entered in Block 2) on dichotomous secondary prevention outcome variables (i.e. taking aspirin, lipid-lowering medication and ACE inhibitors).
A P-value of 
0.01 was set a priori to indicate statistically significant findings. Ethical approval for this study was granted by the Irish College of General Practitioners.
|
Results |
|---|
|
TopAbstractIntroductionMethodologyResultsDiscussionReferences |
|---|
Participants
Fifty-eight practices were asked to participate in the study; thirty-five did so (response rate of 60.3%) and Supplementary Table 3 available online at www.fampr.oxfordjournals.org provides practice descriptive data. A range of methods were used to generate lists of patients, as outlined in Supplementary Table 1 available online at www.fampr.oxfordjournals.org. Data were collected for 1611 patients in total. The mean number of patients recruited from each practice was 46 (range 15–165). Overall practice prevalence of coronary heart disease among GMS eligible patients was 3.16% (95% CI 2.68–3.62). The postal questionnaire was sent to 1577 patients and received from 1084 patients, giving an overall response rate of 68.7%.
The mean age of the sample was 66 (SD 9.1). The gender distribution of the overall sample was 35% female (556) and 65% male (1055). Seventy-nine percent of the overall sample was GMS eligible. According to medical chart review, patients (n = 1610) had made on average four visits (SD 3.6) to their GP in the 6 months prior to the chart search.
The mean time since patients had been first diagnosed with heart disease was 6.92 years (SD 5.8). Fifty-three per cent of patients had angina only; 14% had been diagnosed with an acute MI (AMI) only and 33% with both.
Patient and practice characteristics and adherence to secondary prevention guidelines
Tables 1 and 2 summarize the univariate analyses examining relationships between patients' personal characteristics (gender, type of diagnosis, GMS eligibility, age and time since diagnosis) and adherence to secondary prevention guidelines. These univariate analyses highlight the apparently significant impacts of patient characteristics on the attainment of differing aspects of optimal care. Male patients were significantly more likely than female patients to have cholesterol levels and report exercise levels within the recommended levels; female patients were more likely to have a low fat diet and consume alcohol within the recommended levels. Patients with angina only, compared with patients with a history of MI, CABG or PTCA, were significantly less likely to have been prescribed aspirin, ACE inhibitors, lipid-lowering agents or nitrates, and were less likely to report exercising at the recommended level. GMS eligible patients were significantly less likely than GMS ineligible patients to have been prescribed lipid-lowering agents and were significantly more likely to have been prescribed nitrates. Younger patients were significantly more likely to have been prescribed lipid-lowering agents, less likely to have been prescribed nitrates, more likely to be smokers, more likely to consume alcohol above the recommended level, more likely to be overweight, yet were more likely to engage in exercise at the recommended level.
|
|
Table 3 describes the proportions of patients adhering to secondary prevention guidelines according to practice characteristics. This analysis highlights that practice characteristics appear to have relatively little impact on the attainment of optimal care.
|
Missed opportunities for secondary prevention
Table 4 illustrates the proportions of patients with missed opportunities for secondary prevention.
|
Hierarchical multiple regression analyses
Table 5 shows the results of the hierarchical multiple regressions of selected practice and patient level predictors of blood pressure, cholesterol, exercise score, dietary fat score and number of missed opportunities. The results show that systolic blood pressure was predicted by age only, with older patients having a significantly higher blood pressure (B = 0.31; 95% CI = 0.20–0.43). Cholesterol was predicted by gender only, with females having significantly higher cholesterol (B = –0.41; 95% CI = –0.54 to –0.27). Exercise participation was predicted by age (B = –0.29; 95% CI = –0.45 to –0.13) and gender (B = 5.74; 95% CI = 2.63–8.84); older and female patients exercised less. Dietary fat was predicted by gender; male patients consumed higher levels of dietary fat (B = 4.84; 95% CI = 2.86–6.81). The number of missed opportunities was greater among patients with angina only (B = 0.39; 95% CI = 0.19–0.60). The amount of variance explained by practice and patient variables overall for each of the measures is small, ranging from 2 to 6% of the variance explained in total.
|
Hierarchical logistic regression analyses
Tables 6 shows the hierarchical logistic regressions examining impact of practice variables (entered in Block 1) and patient variables (entered in Block 2) on aspirin, lipid-lowering agent and ACE inhibitor prescription. Having an angina only diagnosis significantly decreased the likelihood of patients being prescribed aspirin (OR = 0.53; 95% CI = 0.40–0.69), lipid-lowering medication (OR = 0.55; 95% CI = 0.43–0.69) or ACE inhibitors (OR = 0.62; 95% CI = 0.48–0.81). Younger patients (OR = 1.05; 95% CI = 1.04–1.06) were also more likely to be prescribed lipid-lowering medication. A longer time elapsed since diagnosis increased the likelihood of being prescribed aspirin (OR = 1.001; 95% CI = 1.001–1.002) or lipid-lowering medication (OR = 1.001; 95% CI = 1.001–1.002). Practice variables had no significant impact on aspirin or statin prescription.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodologyResultsDiscussionReferences |
|---|
The principal findings of this study relate to the impact which the characteristics of individual health systems, providers and patients have on secondary cardiac care. Firstly, the paper is notable in that it reports, from a random group of general practices in a country without universal registration, profiles of patients with established heart disease. The practice prevalence of coronary heart disease for General Medical Services patients was 3.16% (95% CI 2.68–3.62). This compares to 2.3% overall prevalence (95% CI 2.2–2.4) in a similar study carried out in the Grampian region of Scotland.5 The higher crude prevalence reported in our study is to be expected as the GMS figure only includes patients from poorer socioeconomic background. It therefore appears that the methods utilized to identify patients (Supplementary Table 1 is available online at www.fampr.oxfordjournals.org) in a country without universal registration are valid.
Secondly, as found by DeWilde10 practice size or location appears to have little impact on secondary cardiac care. We were somewhat surprised that practice nursing also had no effect. This may reflect the fact that practice lists are generally smaller in Ireland than the UK (Supplementary Table 3 available online at www.fampr.oxfordjournals.org), perhaps resulting in a more direct role for practitioners in chronic disease management. This however requires further confirmatory work.
Interpretation of the impact of personal characteristics is more complex. The impacts of individual characteristics, such as gender or socioeconomic status, were attenuated in the regression analyses suggesting possible confounding. The most consistent significant finding was that patients with a diagnosis of angina only were significantly less likely to receive aspirin, statins or ACE inhibitors and more likely to have more missed opportunities. Reviewing gender and statin use only, our findings, with no inequality being found, were similar to those of DeWilde10 and Reid,13 but not Hippisley-Cox.11 We found no association between gender and aspirin or ACE inhibitor usage or the number of missed opportunities. As described by DeWilde,10 Reid13 and Majeed,12 we found that older patients were less likely to be prescribed statins. As the absolute risk of further cardiovascular events increases with age, this merits consideration at local practice level. Performing socioeconomic assessment at the individual level (GMS eligibility), we found no associations with medical or lifestyle management which broadly confirms the findings of Reid13 and DeWilde10 which were restricted to statin usage only. In the provision of secondary cardiac care at the practice level, special consideration should be therefore afforded to older patients with a diagnosis of angina only.
Due to the random selection of practices, good practice and patient response rates, use of individual data and large sample size, high external validity can be expected. Australia and the United States both have health systems without universal registration. Our findings regarding the quality of secondary cardiac care are similar to those of the large US Third National Health and Nutrition Examination Survey.18 In Australia, similar community-based profiles of secondary cardiac care are not yet available (Stocks N, personal communication). However the baseline results of patients recruited for a hospital-based intervention were also broadly similar.19 Of interest is that the profiles for the main outcome measurements in these three countries were broadly similar to those in the UK Heart and Health Study cohort immediately before the implementation of the National Service Framework.20 The UK Quality and Outcomes Framework is a unique source for the provision of national data regarding the quality of secondary cardiac care in the community (http://www.ic.nhs.uk/services/qof/data/; accessed 1 December 2006). How health systems without universal registration will produce equivalent data remains to be seen.
Study methodology
There are a number of potential limitations to the study. Firstly, as it is based in practices on the western seaboard of Ireland application to large urban centres may be problematical. The study involved review of practice charts. We accept that charts may be a flawed record of what has occurred during consultations and may be an underestimation of what information is actually available to the practitioner. Pragmatically however, the use of practice charts appears to be an appropriate and reasonable approach which facilitates comparison with other work. Over the counter usage of aspirin was also not recorded.
The use of general practice record data for individuals is a particular strength of the study thereby avoiding the risk of ecological fallacy. Of similar papers, DeWilde,10 Hippisley-Cox,11 Majeed12 and Carroll9 all utilized large routinely collected medical databases. Reid13 used a community-based survey of 760 patients—the Health Survey for England. With the exception of Hippisley-Cox,11 our study extends such previous work which was restricted to statin usage only. None, apart from DeWilde,10 included practice size or location.
The Joint European Task Force recommendations14 state that the community needs to provide continuity of risk-factor management and to ensure long-term compliance with evidence-based therapies. This study records that the provision of general practice secondary cardiac care in Irish general practice is not optimal. However, it is important to acknowledge the significant amount which has been achieved in a health system accepted as not being conducive to the provision of chronic medical care within the community. The opportunity for potential health gain is highlighted. In Ireland, a national pilot programme for the provision of secondary care (‘Heartwatch’) has recently been established to achieve such gain.21 The EUROASPIRE studies22 are significant in the application of similar methodology to incident patients identified in hospitals in 15 European countries. We suggest that a similar study based in European general practice would be both rewarding and opportune.
|
Acknowledgments |
|---|
We are very grateful to Drs Sean Denyer, Peter Wright and Mary Hynes of the North Western and Western Health Boards for their encouragement and support. We wish to acknowledge the funding received from these two boards together with the Health Research Board Health Services Research Fellowship awarded to Ms Molly Byrne. The study would not have been possible without the huge interest and support of the participating GPs, who are listed separately. This work was completed whilst AWM was on sabbatical at the Departments of General Practice at Flinders University and the University of Adelaide, Australia.
Participating GPs (One GP requested to remain anonymous):
Drs John O' Sullivan; Richard Joyce; Marcus Allen; Patrick O' Malley; Martin Daly; Desmond Bluett; Seamus Cryan; Bernard McGuire; Declan Clinton; Marian Brogan; Richard Tobin; Michael Regan; David Townley; Kay Moran; Enda Harhen; Kieran Whyte; Daniel Murphy; Kieran O' Reilly; Edward Harty; Vivian Brennan; Neil Farrell; John Sheerin; Brendan Forkan; Pauric Mitchell; Ciaran Kelly; John-Mark Dick; Anthony Delap; Sean Bourke; James Brogan; Paul Money; Charles Bourke; Roddy Quinn; Eamonn, John Shea and Regan; Margaret Gilligan King.
Competing interests: AWM has received funding from Pfizer to support educational meetings for GPs who teach medical students from the Department of General Practice at NUI, Galway. There are no other competing interests.
|
Notes |
|---|
Byrne M, Murphy AW, Walsh JC, Shryane E, McGroarty M and Kelleher CC. A cross-sectional study of secondary cardiac care in general practice: Impact of personal and practice characteristics. Family Practice 2006; 23: 295–302.
|
References |
|---|
|
TopAbstractIntroductionMethodologyResultsDiscussionReferences |
|---|
1 McAlister F, Lawson F, Teo K, Armstrong P. (2001) Randomised trials of secondary prevention programmes in coronary heart disease: systematic review. BMJ 323:957–962.
2 Bradley F and Cupples ME. (1999) Reducing the risk of recurrent coronary heart disease. BMJ 318:1499–1500.
3 Faergeman O. (2001) A collective failure of medical practice? Eur Heart J 22:526–528.
4 Starfield B. (1998) Primary Care: Balancing Health Needs, Services and Technology 2nd edn. (Oxford University Press, Oxford).
5 Campbell NC, Ritchie LD, Thain J, Deans HG, Rawles JM, Squair JL. (1998) Secondary prevention in coronary heart disease: a randomised trial of nurse led clinics in primary care. Heart 80:447–452.
6 Jolly K, Bradley F, Sharp S, et al. (1999) Randomised controlled trial of follow up care in general practice of patients with myocardial infarction and angina: final results of the Southampton heart integrated care project (SHIP). The SHIP Collaborative Group. BMJ 318:706–711.
7 Campbell NC, Thain J, Deans HG, Ritchie LD, Rawles JM. (1998) Secondary prevention in coronary heart disease: baseline survey of provision in general practice. BMJ 316:1430–1434.
8 Moher M, Yudkin P, Wright L, et al. (2001) Cluster randomised controlled trial to compare three methods of promoting secondary prevention of coronary heart disease in primary care. BMJ 322:1338.
9 Carroll K, Majeed A, Firth C, Gray J. (2003) Prevalence and management of coronary heart disease in primary care: population-based cross-sectional study using a disease register. J Public Health Med 25:29–35.
10 deWilde S, Carey IM, Bremner SA, Richards N, Hilton SR, Cook DG. (2003) Evolution of statin prescribing 1994–2001: a case of agism but not of sexism? Heart 89:417–421.
11 Hippisley-Cox J, Pringle M, Crown N, Meal A, Wynn A. (2001) Sex inequalities in ischaemic heart disease in general practice: cross sectional survey. BMJ 322:832.
12 Majeed A, Moser K, Maxwell R. (2000) Age, sex and practice variations in the use of statins in general practice in England and Wales. J Public Health Med 22:275–279.
13 Reid FD, Cook DG, Whincup PH. (2002) Use of statins in the secondary prevention of coronary heart disease: is treatment equitable? Heart 88:15–19.
14 Wood D, DeBacker G, Faergeman O, Graham I, Mancia G, Pyorala K. (1998) Prevention of coronary heart disease in clinical practice. Recommendations of the Second Joint Task Force of European and other Societies on coronary prevention. Eur Heart J 19:1434–1503.
15 Byrne M, Walsh J, Murphy A. (2005) Secondary prevention of coronary heart disease: patient beliefs and health-related behaviour. J Psychosom Res 58:403–415.[CrossRef][Web of Science][Medline]
16 Godin G and Shephard R. (1985) A simple method to assess exercise behavior in the community. Can J Appl Sport Sci 10:141–146.[Medline]
17 Roe L, Strong C, Whiteside C, Neil A, Mant D. (1994) Dietary intervention in primary care: validity of the DINE method for diet assessment. Fam Pract 11:375–381.
18 Qureshi AI, Suri FK, Guterman LR, Hopkins N. (2001) Ineffective secondary prevention in survivors of cardiovascular events in the US population. Arch intern Med 161:1621–1628.
19 Vale MJ, Jelinek MV, Best JD, et al. (2003) Coaching patients On Achieving Cardiovascular Health (COACH): a multicenter randomized trial in patients with coronary heart disease. Arch Intern Med 163:2775–2783.
20 Lawlor DA, Whincup P, Emberson JR, Rees K, Walker M, Ebrahim S. (2004) The challenge of secondary prevention for coronary heart disease in older patients: findings from the British Women's Heart and Health Study and the British Regional Heart Study. Fam Pract 21:582–586.
21 Heartwatch Steering Committee. HeartWatch Annual Report; Dublin. 2004. http://www.icgp.ie/index.cfm/loc/6-4.htm (accessed 12 October 2005).
22 Euroaspire I, Group II, European Action on Secondary Prevention by Intervention to Reduce E. (2001) Clinical reality of coronary prevention guidelines: a comparison of EUROASPIRE I and II in nine countries. EUROASPIRE I and II Group. European Action on Secondary Prevention by Intervention to Reduce Events. Lancet 357:995–1001.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  B Buckley and A W Murphy Do patients with angina alone have a more benign prognosis than patients with a history of acute myocardial infarction, revascularisation or both? Findings from a community cohort study Heart, March 1, 2009; 95(6): 461 - 467. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. G. Glynn, D. Reddan, J. Newell, J. Hinde, B. Buckley, and A. W. Murphy Chronic kidney disease and mortality and morbidity among patients with established cardiovascular disease: a West of Ireland community-based cohort study Nephrol. Dial. Transplant., September 1, 2007; 22(9): 2586 - 2594. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Buckley, A. W Murphy, M. Byrne, and L. Glynn Selection bias resulting from the requirement for prior consent in observational research: a community cohort of people with ischaemic heart disease Heart, September 1, 2007; 93(9): 1116 - 1120. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||