Cardiovascular risk prediction in type 2 diabetes before and after widespread screening: a derivation and validation study.

BACKGROUND: Until recently, most patients with diabetes worldwide have been diagnosed when symptomatic and have high cardiovascular risk, meaning most should be prescribed cardiovascular preventive medications. However, in New Zealand, a world-first national programme led to approximately 90% of eligible adults being screened for diabetes by 2016, up from 50% in 2012, identifying many asymptomatic patients with recent-onset diabetes. We hypothesised that cardiovascular risk prediction equations derived before widespread screening would now significantly overestimate risk in screen-detected patients. METHODS: New Zealanders aged 30-74 years with type 2 diabetes and without known cardiovascular disease, heart failure, or substantial renal impairment were identified from the 400 000-person PREDICT primary care cohort study between Oct 27, 2004, and Dec 30, 2016, covering the period before and after widespread screening. Sex-specific equations estimating 5-year risk of cardiovascular disease were developed using Cox regression models, with 18 prespecified predictors, including diabetes-related and renal function measures. Equation performance was compared with an equivalent equation derived in the New Zealand Diabetes Cohort Study (NZDCS), which recruited between 2000 and 2006, before widespread screening. FINDINGS: 46 652 participants were included in the PREDICT-1° Diabetes subcohort, of whom 4114 experienced first cardiovascular events during follow-up (median 5·2 years, IQR 3·3-7·4). 14 829 (31·8%) were not taking oral hypoglycaemic medications or insulin at baseline. Median 5-year cardiovascular risk estimated by the new equations was 4·0% (IQR 2·3-6·8) in women and 7·1% (4·5-11·2) in men. The older NZDCS equation overestimated median cardiovascular risk by three times in women (median 14·2% [9·7-20·0]) and two times in men (17·1% [4·5-20·0]). Model and discrimination performance measures for PREDICT-1° Diabetse equations were also significantly better than for the NZDCS equation (eg, for women: R2=32% [95% CI 29-34], Harrell's C=0·73 [0·72-0·74], Royston's D=1·410 [1·330-1·490] vs R2=24% [21-26], C=0·69 [0·67-0·70], and D=1·147 [1·107-1·187]). INTERPRETATION: International treatment guidelines still consider most people with diabetes to be at high cardiovascular risk; however, we show that recent widespread diabetes screening has radically changed the cardiovascular risk profile of people with diabetes in New Zealand. Many of these patients have normal renal function, are not dispensed glucose-lowering medications, and have low cardiovascular risk. These findings have clear international implications as increased diabetes screening is inevitable due to increasing obesity, simpler screening tests, and the introduction of new-generation glucose-lowering medications that prevent cardiovascular events. Cardiovascular risk prediction equations derived from contemporary diabetes populations, with multiple diabetes-related and renal function predictors, will be required to better differentiate between low-risk and high-risk patients in this increasingly heterogeneous population and to inform appropriate non-pharmacological management and cost-effective targeting of expensive new medications. FUNDING: Health Research Council of New Zealand, Heart Foundation of New Zealand, and Healthier Lives National Science Challenge.

Retrospective review of abdominal aortic aneurysm deaths in New Zealand: what proportion of deaths is potentially preventable by a screening programme in the contemporary setting?

OBJECTIVES: To describe the proportions of people dying from abdominal aortic aneurysm (AAA) who might have benefited from a formal screening programme for AAA. DESIGN: Retrospective cross-sectional review of deaths. SETTING AND STUDY POPULATIONS: All AAA deaths registered in New Zealand from 2010 to 2014 in the absence of a national AAA screening programme. MAIN OUTCOME MEASURES: Known history of AAA prior to the acute event leading to AAA death, prognosis limiting comorbidities, history of prior abdominal imaging and a validated multimorbidity measure (M3-index scores). RESULTS: 1094 AAA deaths were registered in the 5 years between 2010 and 2014 in New Zealand. Prior to the acute AAA event resulting in death, 31.3% of the cohort had a known AAA diagnosis, and 10.9% had a previous AAA procedure. On average, the AAA diagnosis was known 3.7 years prior to death. At least 77% of the people dying from AAA also had one or more other prognosis limiting diagnosis. The hazard of 1-year mortality associated with the non-AAA related comorbidities for the AAA cohort aged 65 or above were 1.5-2.6 times higher than to the age matched general population based on M3-index scores. In 2014, overall AAA deaths accounted for only 0.7% of total deaths, and 1.0% of deaths among men aged 65 or above in New Zealand. At most, 20% of people dying from AAA in New Zealand between 2010 and 2014 might have had the potential to derive full benefit from a screening programme. About 51% of cases would have derived no or very limited benefit from a screening programme. CONCLUSION: Falling AAA mortality, and high prevalence of competing comorbidities and/or prior AAA diagnosis and procedure raises the question about the likely value of a national AAA screening programme in a country such as New Zealand.

Is general practice identification of prior cardiovascular disease at the time of CVD risk assessment accurate and does it matter?

AIMS: To determine the accuracy of general practice recording of prior cardiovascular disease (CVD) at the time of CVD risk assessment and whether recording impacts on CVD management. METHODS: Prior CVD status entered at the time of a first CVD risk assessment from 2002-2015 was compared to prior ischaemic CVD hospitalisations from national datasets using anonymous linkage with an encrypted National Health Index identifier. Clinical factors associated with inaccurate recording of prior events were identified using multivariable logistic regression. The impact of recording accuracy was assessed by the dispensing of CVD preventive medications in the six months after first CVD risk assessment. RESULTS: Among 454,369 people aged 35-74 years who had CVD risk assessments, 30,924 (6.8%) had previously been admitted with ischaemic CVD. Of these people, only 61% were recorded as having prior CVD during risk assessment, with better recording for coronary and stroke events than for peripheral vascular procedures. Inaccurate primary care recording was more likely for younger people (<55 years), women, Maori, Pacific, Indian and Asian ethnic groups whereas smokers and people with diabetes were more likely to have prior CVD correctly identified. Over more than a decade, the odds of inaccurate recording during risk assessment increased [OR 1.09 (95% CIs 1.08-1.10)]. If prior CVD was entered at the time of risk assessment then dispensing of blood pressure-lowering, lipid-lowering, antiplatelet/anticoagulant medications, separately or together, was higher (86%, 85%, 83% and 69%, respectively) than if not recorded (70%, 60%, 60% and 43%). CONCLUSIONS: Overall, 39% of people with prior CVD hospitalisations were not recorded as having prior CVD when their CVD risk was first assessed in general practice. This was associated with inequities in evidence-based risk management. System-based measures are required for robust data sharing at the time of clinical decision making.

Can administrative health utilisation data provide an accurate diabetes prevalence estimate for a geographical region?

AIM: To validate the New Zealand Ministry of Health (MoH) Virtual Diabetes Register (VDR) using longitudinal laboratory results and to develop an improved algorithm for estimating diabetes prevalence at a population level. METHODS: The assigned diabetes status of individuals based on the 2014 version of the MoH VDR is compared to the diabetes status based on the laboratory results stored in the Auckland regional laboratory result repository (TestSafe) using the New Zealand diabetes diagnostic criteria. The existing VDR algorithm is refined by reviewing the sensitivity and positive predictive value of the each of the VDR algorithm rules individually and as a combination. RESULTS: The diabetes prevalence estimate based on the original 2014 MoH VDR was 17% higher (n = 108,505) than the corresponding TestSafe prevalence estimate (n = 92,707). Compared to the diabetes prevalence based on TestSafe, the original VDR has a sensitivity of 89%, specificity of 96%, positive predictive value of 76% and negative predictive value of 98%. The modified VDR algorithm has improved the positive predictive value by 6.1% and the specificity by 1.4% with modest reductions in sensitivity of 2.2% and negative predictive value of 0.3%. At an aggregated level the overall diabetes prevalence estimated by the modified VDR is 5.7% higher than the corresponding estimate based on TestSafe. CONCLUSION: The Ministry of Health Virtual Diabetes Register algorithm has been refined to provide a more accurate diabetes prevalence estimate at a population level. The comparison highlights the potential value of a national population long term condition register constructed from both laboratory results and administrative data.

First and recurrent ischaemic heart disease events continue to decline in New Zealand, 2005-2015.

OBJECTIVES: To examine recent trends in first and recurrent ischaemic heart disease (IHD) deaths and hospitalisations. METHODS: Using anonymous patient-linkage of routinely collected data, all New Zealanders aged 35-84 years who experienced an International Statistical Classification of Diseases and Related Health Problems I(CD)-coded IHD hospitalisation and/or IHD death between 1 January 2005 and 31 December 2015 were identified. A 10-year look-back period was used to differentiate those experiencing first from recurrent events. Age-standardised hospitalisation and mortality rates were calculated for each calendar year and trends compared by sex and age. RESULTS: 160 109 people experienced at least one IHD event (259 678 hospitalisations and 35 548 deaths) over the 11-year study period, and there was a steady decline in numbers (from almost 24 000 in 2005 to just over 16 000 in 2015) and in age-standardised rates each year. With the exception of deaths in younger (35-64 years) women with prior IHD, there was a significant decline in IHD events in men and women of all ages, with and without a history of IHD. The decline in IHD mortality was greater for those experiencing a first rather than recurrent IHD event (3.8%-5.2% vs 0%-3.7% annually on average). In contrast, the decline in IHD hospitalisations was greater for those experiencing a recurrent compared with a first IHD event (5.6%-7.3% vs 3.2%-5.7% annually on average). CONCLUSIONS: The substantial decline in IHD hospitalisations and mortality observed in New Zealanders with and without prior IHD between 2005 and 2015 suggests that primary and secondary prevention efforts have been effective in reducing the occurrence of IHD events.

Service planning implications of estimating Primary Health Organisation enrolment rate based on a Health Service Utilisation population rather than a Census-derived population.

AIM: Estimating Primary Health Organisation (PHO) enrolment rates with a census-derived estimated resident population denominator may provide misleading results because of numerator and denominator mismatch. This study uses the Health Service Utilisation (HSU) population denominator as an alternative. METHOD: A HSU population was generated by record linkage of routinely collected datasets from the Ministry of Health via encrypted National Health Index (NHI). We compare PHO enrolment rates by age and ethnicity in Counties Manukau District Health Board (CMDHB) in 2013. RESULTS: In CMDHB, 98% of people who had utilised publicly-funded health services in 2013 were enrolled in a PHO in 2013. Using the HSU population as a denominator, PHO enrolment rates for Maaori, Pacific, Asian, New Zealand European/Other population groups were 98.3%, 97.7%, 97.6%, and 98.3% respectively. Just under 4% of people discharged from CMDHB inpatient facilities were not enrolled in a PHO within a month from the day of discharge in 2013. CONCLUSION: Using the HSU population as a proxy of health services need, PHO enrolment rates were similar across ethnicities in the CMDHB population. Support to improve PHO enrolment coverage would be more efficient if the HSU population were used to target people who are not yet enrolled in a PHO.

The future of population registers: linking routine health datasets to assess a population's current glycaemic status for quality improvement.

OBJECTIVES: To determine the diabetes screening levels and known glycaemic status of all individuals by age, gender and ethnicity within a defined geographic location in a timely and consistent way to potentially facilitate systematic disease prevention and management. DESIGN: Retrospective observational study. SETTING: Auckland region of New Zealand. PARTICIPANTS: 1 475 347 people who had utilised publicly funded health service in New Zealand and domicile in the Auckland region of New Zealand in 2010. The health service utilisation population was individually linked to a comprehensive regional laboratory repository dating back to 2004. OUTCOME MEASURES: The two outcomes measures were glycaemia-related blood testing coverage (glycated haemoglobin (HbA1c), fasting and random glucose and glucose tolerance tests), and the proportions and number of people with known dysglycaemia in 2010 using modified American Diabetes Association (ADA) and WHO criteria. RESULTS: Within the health service utilisation population, 792 560 people had had at least one glucose or HbA1c blood test in the previous 5.5 years. Overall, 81% of males (n=198 086) and 87% of females (n=128 982) in the recommended age groups for diabetes screening had a blood test to assess their glycaemic status. The estimated age-standardised prevalence of dysglycaemia was highest in people of Pacific Island ethnicity at 11.4% (95% CI 11.2% to 11.5%) for males and 11.6% (11.4% to 11.8%) for females, followed closely by people of Indian ethnicity at 10.8% (10.6% to 11.1%) and 9.3% (9.1% to 9.6%), respectively. Among the indigenous Maori population, the prevalence was 8.2% (7.9% to 8.4%) and 7% (6.8% to 7.2%), while for 'Others' (mainly Europeans) it was 3% (3% to 3.1%) and 2.2% (2.1% to 2.2%), respectively. CONCLUSIONS: We have demonstrated that the data linkage between a laboratory repository and national administrative datasets has the potential to provide a systematic and consistent individual level clinical information that is relevant to medical auditing for a large geographically defined population.

Initiation and maintenance of cardiovascular medications following cardiovascular risk assessment in a large primary care cohort: PREDICT CVD-16.

AIM: To examine whether use of a standardized cardiovascular disease (CVD) risk assessment recommended by national guidelines is associated with appropriate initiation and maintenance of medication in a large primary care cohort. METHODS AND DESIGN: A total of 90,631 people aged 30-80 years were followed for up to 3 years after a formal CVD risk assessment was undertaken between January 2006 and October 2009, during routine primary care visits in New Zealand. Patients either had prior CVD or had their CVD risk estimated using a modified Framingham prediction equation for fatal or non-fatal CVD events. The individual risk profiles were anonymously linked to national dispensing data for blood-pressure-lowering and lipid-lowering medications in the 6-month period before and in consecutive 6-month blocks after the baseline CVD risk assessment. RESULTS: At baseline, a combination of blood-pressure-lowering and lipid-lowering therapy was already being used by about two-thirds of patients with prior CVD, one-quarter with a 5-year CVD risk greater than 10% (approximately 20% 10-year risk), and one-tenth with CVD risk below this level. Among these previously treated patients, dispensing rates for blood-pressure-lowering, lipid-lowering, or both medications together declined by only 4⊟16% up to 3 years after baseline assessment, irrespective of risk category. Among patients untreated at baseline, combination therapy was initiated within 6 months for 21% with prior CVD, 16% with 5-year CVD risk greater than 15% (approximately 30% 10-year risk and the national drug-treatment threshold), 10% with 5-year CVD risk between 10 and 14% (approximately 20⊟29% 10-year risk), and 3% in the lowest risk category. Across the study population, patients with prior CVD had the highest dispensing rates for each category of medication, and incrementally higher dispensing rates were noted as CVD risk group increased. CONCLUSIONS: In this primary care cohort, most patients already using CVD medications at the time of the baseline CVD risk assessment maintained treatment over a maximum of 3 years follow up, irrespective of their estimated baseline risk. Among patients untreated at baseline, subsequent dispensing rates were strongly related to estimated CVD risk group. Around 15⊟20% of untreated patients meeting national drug-treatment criteria commenced combination pharmacotherapy within 6 months of CVD risk assessment.

Statin use in COPD patients is associated with a reduction in mortality: a national cohort study.

AIMS: To assess whether statin use is associated with reduced mortality in patients with chronic obstructive pulmonary disease (COPD). METHODS: Hospitalisation, drug dispensing, and mortality records were linked for New Zealanders aged 50-80 years discharged from hospital with a first admission with COPD in 2006. Patients were classified according to whether or not they were prescribed statins prior to admission. Baseline characteristics were compared and hazard ratios calculated for statin users versus statin non-users for all-cause mortality over follow-up of up to 4 years. RESULTS: A total of 1,687 patients (mean age 70.6 years) were followed, including 596 statin users and 1,091 non-users. There were more men in the statin user group (58.4% vs. 48.5%), and statin users were more likely to have a history of cardiovascular disease (58.6% vs. 25.1%), prescription for frusemide as a proxy for heart failure (47.7% vs. 24.5%) or diabetes (35.4% vs.11.6%) than statin non-users (p<0.001). A total of 671 deaths occurred during the follow-up period. After adjustment for age, sex, ethnic group, history of cardiovascular disease, diabetes, and prescription for frusemide, the hazard ratio for statin users vs. statin non-users for all-cause mortality was 0.69 (95% CI 0.58 to 0.84). CONCLUSIONS: Statin use is associated with a 30% reduction in all-cause mortality at 3-4 years after first admission for COPD, irrespective of a past history of cardiovascular disease and diabetes.

Four out of ten patients are not taking statins regularly during the 12 months after an acute coronary event.

BACKGROUND: In New Zealand, a setting in which national guidelines recommend statins for all patients with coronary heart disease (CHD) and cost barriers are low, patterns of use of these drugs are unknown. We investigated dispensing rates after hospital discharge for acute CHD event. DESIGN: Retrospective cohort study. METHODS: Drug dispensing, hospital diagnosis, and mortality records were linked by unique identifier for all New Zealanders aged 35-84 years after discharge following acute CHD event in 2007. We defined the statin dispensing ratio (SDR) as the proportion of days that 15,506 patients aged 35-84 years were dispensed such agents during the 12 months post discharge. An SDR ≥0.8 (80% or more days covered) was considered optimal. RESULTS: Overall, 59% of the cohort had an SDR ≥0.8. Of patients dispensed statins in the 3 months before admission (n = 5506), almost all (99%; 5466) continued treatment during follow up and 82% had an SDR ≥0.8. In contrast, for patients not dispensed statins before admission (n = 8014), only two-thirds started statins during follow up and only 44% had an SDR ≥0.8. Of all patients with low statin dispensing (SDR <0.8), about one-quarter were not dispensed any lipid-lowering drugs, one-quarter received alternative lipid-lowering drugs, one-quarter stopped statins, and the remaining quarter were intermittent statin users. CONCLUSION: In a setting with few barriers to statin treatment, about 40% of patients had suboptimal statin dispensing during the year after hospital treatment for CHD. This study has identified four significant categories of suboptimal adherence that could inform quality improvement programmes.

Can the prevalence of diagnosed diabetes be estimated from linked national health records? The validity of a method applied in New Zealand.

INTRODUCTION: With projected global increases in the prevalence of Type 2 diabetes, the health sector requires timely assessments of the prevalence of this disease to monitor trends, plan services, and measure the efficacy of prevention programmes. AIM: To assess the validity of a method to estimate the prevalence of diagnosed diabetes from linked national health records. METHODS: We measured the agreement between a diabetes diagnosis (using combined national lists of drug dispensing, outpatient attendance, laboratory tests (HbA1c) and hospital diagnoses) and a primary care diabetes diagnosis in a (PREDICT™) cohort of 53,911 adult New Zealanders. The completeness of the diagnosis of diabetes in the cohort was estimated using capture-recapture methods. RESULTS: The primary care cohort had a high prevalence of recorded diabetes (20.9%, 11,266/53,911), similar to our derived prevalence of 20.1%. Of the participants with a diagnosis of diabetes, 89% (10,182/11,266) had a similar derived diagnosis, indicating that only about one in 10 people with a primary care diagnosis had not been either admitted to hospital, seen at outpatient clinics, prescribed diabetes drugs or undertaken regular HbA1c tests. The capture-recapture prevalence of diagnosed diabetes in this cohort was 23.7% indicating that primary care diagnoses in the cohort were about 90% complete. DISCUSSION: A method for estimating the prevalence of diagnosed diabetes from national health data shows high-level agreement with primary care records. Linked health data can provide an efficient method for estimating the prevalence of diagnosed diabetes in regions where such records are individually linked.

Sociodemographic differences in prevalence of diagnosed coronary heart disease in New Zealand estimated from linked national health records.

AIM: To estimate sociodemographic differences in the prevalence of coronary heart disease (CHD) in New Zealand from linked health records. METHODS: We combined records of hospital treatment for CHD, dispensing of selected anti-anginal drugs and mortality to estimate the national point prevalence of coronary heart disease in New Zealand in December 2008. Stratified estimates are presented by gender; age; Maori, Pacific, Indian and 'Other' (mainly New Zealand European) ethnic groups; and socioeconomic status. RESULTS: Among a "health contact" population of adults (greater than and equal to 15 years), about one in twenty (6.5% of men and 4.1% of women) had indicators of a past diagnosis or treatment for CHD or both. Substantial differences in prevalence occurred by gender, ethnic group and socioeconomic status. For example, among New Zealanders aged 35 to 74 years, Indian men had the highest age-adjusted prevalence (7.78%; 95%CI 7.43 to 8.15), almost double the prevalence of 'Other' males. Among women, Maori had the highest adjusted prevalence (4.03%; 95% CI 3.89 to 4.17), just over twice that of 'Others.' CONCLUSION: Major sociodemographic disparities in the national burden of CHD persist. Our results are similar to previous studies of ethnic disparities in CHD incidence, but also confirm concerns about the emerging CHD burden among South Asians. Indian males have the highest CHD prevalence of any gender-specific ethnic group. Of equal concern, Maori women have a similar prevalence to European males.

Healthcare services funded by Counties Manukau District Health Board for people in the last year of life.

INTRODUCTION: The last year of life is often associated with a high level of healthcare utilisation and cost. To date, little information is available regarding the healthcare utilisation patterns in the last year of life in New Zealand. AIM: To describe the healthcare utilisation patterns and costs of the residents of Counties Manukau District Health Board (CMDHB) region in the 1-year period prior to death in 2008. METHOD: CMDHB residents who died in 2008 were identified from the National Mortality Dataset. The health services utilisation patterns and costs in the last year of life were derived from National Minimum Dataset (NMDS), Pharmaceutical Collection, Laboratory Claims Collection, and National Non-Admitted Patient Collection via encrypted NHI linkage. RESULTS: Forty percent of all deaths in 2008 in CMDHB occurred in a publicly funded hospital. Just over 80% of people had at least one inpatient hospital stay in the last year of life. More than 75% of the healthcare costs funded by CMDHB in the last year of life were related to inpatient hospitalisations. The average cumulative length of inpatient stay over the year in the people who had an inpatient event was 20.6 days. Outpatient, pharmaceutical, and laboratory services were received by 84%, 91%, and 86% of people respectively in their last year of life. CONCLUSION: Consistent with the international literature, this study found that CMDHB residents in the last year of life have a high level of health service utilisation. Decisions about the appropriate use of high cost health services in people towards the end of life can be extremely challenging. These decisions are resource allocation decisions as well as clinical decisions and should be based on clinical factors, cost utilities, and patient, family, and society's expectations.

Under-utilisation of preventive medication in patients with cardiovascular disease is greatest in younger age groups (PREDICT-CVD 15).

INTRODUCTION: Blood pressure-lowering (BPL) and lipid-lowering (LL) medications together reduce estimated absolute five-year cardiovascular disease (CVD) risk by >40%. International studies indicate that the proportion of people with CVD receiving pharmacotherapy increases with advancing age. AIM: To compare BPL and LL medications, by sociodemographic characteristics, for patients with known CVD in primary care settings. METHODS: The study population included patients aged 35-74 with known CVD assessed in primary care from July 2006 to October 2009 using a web-based computerised decision support system (PREDICT) for risk assessment and management. Clinical data linked anonymously to national sociodemographic and pharmaceutical dispensing databases. Differences in dispensing BPL and LL medications in six months before first PREDICT assessment was analysed according to age, sex, ethnicity and deprivation. RESULTS: Of 7622 people with CVD, 1625 <55 years old, 2862 were women and 4609 lived in deprived areas (NZDep quintiles 4/5). The study population included 4249 European, 1556 Maori, 1151 Pacific and 329 Indian peoples. BPL medications were dispensed to 81%, LL medications to 73%, both BPL and LL medications to 67%, and 87% received either class of medication. Compared with people aged 65-75, people aged 35-44 were 30-40% less likely and those aged 45-54 were 10-15% less likely to be dispensed BPL, LL medications or both. There were minimal differences in likelihood of dispensing according to sex, ethnicity or deprivation. DISCUSSION: BPL and LL medications are under-utilised in patients with known CVD in New Zealand. Only two-thirds of patients in this cohort are on both. Younger patients are considerably less likely to be on recommended medications.

A population-based approach to the estimation of diabetes prevalence and health resource utilisation.

AIM: This study estimated diabetes prevalence and utilisation of healthcare services in Counties Manukau using routinely collected administrative data and compared estimates with findings for three other district health boards (DHBs) in close geographic proximity. METHOD: Records of subsidy claims for pharmaceuticals and laboratory investigations were linked to records in a national hospital admissions database to 'reconstruct' populations of four DHBs--Counties Manukau, Northland, Waitemata and Auckland. Individuals were included in reconstructed populations if they had health events recorded between January 2006 and December 2007. Diabetes cases were identified using an algorithm based on claims for monitoring tests and pharmaceuticals, as well as clinical codes for diabetes in hospital admissions. RESULTS: Reconstructed populations were only 6% lower than census population counts indicating that the vast majority of the population use health services in a two year period. The age- and sex-standardised prevalence of diabetes was 7.1% in Counties Manukau and 5.2% in the other three DHBs combined. Prevalence of diabetes was highest amongst Maori (10.6% in women and 12.2% in men) and Pacific peoples (15.0% for women and 13.5% for men). Maori diabetes cases had the highest hospital discharge rate of any ethnic group. Community pharmaceutical prescribing patterns and laboratory test frequency were similar between diabetes cases by ethnicity and deprivation. CONCLUSION: Estimates of diabetes prevalence using linkage of routinely collected administrative data were consistent with epidemiological surveys, suggesting that linkage of pharmaceutical and laboratory subsidy databases with hospital admissions data can be used as an alternative to traditional surveys for estimating the prevalence of some long-term conditions. This study demonstrated substantial differences in the prevalence of diabetes and in hospitalisation rates by ethnicity, but measures of community diabetes care were similar by ethnicity and deprivation.

Estimating diabetes prevalence in South Auckland: how accurate is a method that combines lists of linked health datasets?

AIMS: To assess the accuracy of a method for estimating adult diabetes prevalence that combines linked, routine health datasets in South Auckland, New Zealand. METHODS: We used a simple algorithm that combined records of laboratory testing, drug dispensing and hospital diagnoses applied to National Health Index-linked health data in South Auckland to estimate the prevalence of diabetes in 2007. We investigated the sensitivity of this 'combined list' algorithm against a gold standard of individuals with diagnosed diabetes enrolled in a Chronic Care Management programme (CCMP). We also assessed the sensitivity of this algorithm against an estimated diabetes population generated using capture-recapture methods. RESULTS: From the combined-list algorithm, 25,797 (7.2%) South Aucklanders aged 15 years and over had diabetes. During this period, 10,725 patients were enrolled in the CCMP. The combined list algorithm correctly identified (sensitivity) 10,351/10,725 (96.5%) of those enrolled. When we used the capture-recapture estimated diabetes population as an alternative gold standard, 34,418 [9.5%] of South Aucklanders 15 years and over had diabetes, with the sensitivity of the combined list method falling to about 75% (25,797/34,418). CONCLUSION: Linked health data provide reasonably accurate estimates of diabetes prevalence in a New Zealand population; particularly for cases with longstanding or complicated disease.

The burden of coronary heart disease in Maori: population-based estimates for 2000-02.

OBJECTIVE: To estimate coronary heart disease (CHD) incidence, prevalence, survival, case fatality and mortality for Maori, in order to support service planning and resource allocation. METHODS: Incidence was defined as first occurrence of a major coronary event, i.e. the sum of first CHD hospital admissions and out-of-hospital CHD deaths in people without a hospital admission for CHD in the preceding five years. Data for the years 2000-02 were sourced from the New Zealand Health Information Service and record linkage was carried out using a unique national identifier, the national health index. RESULTS: Compared to the non-Maori population, Maori had both elevated CHD incidence and higher case fatality. Median age at onset of CHD was younger for Maori, reflecting both higher age specific risks and younger population age structure. The lifetable risk of CHD for Maori was estimated at 37% (males) and 34% (females), only moderately higher than the corresponding estimates for the non-Maori population, despite higher Maori CHD incidence. This reflects the offsetting effect of the higher 'other cause' mortality experienced by Maori. Median duration of survival with CHD was similar to that of the non-Maori population for Maori males but longer for Maori females, which is most likely related to the earlier age of onset. CONCLUSIONS: This study has generated consistent estimates of CHD incidence, prevalence, survival, case fatality and mortality for Maori in 2000-02. The inequality identified in CHD incidence calls for a renewed effort in primary prevention. The inequality in CHD case fatality calls for improvement in access for Maori to secondary care services.

Ethnic and socioeconomic disparities in the prevalence of cardiovascular disease in New Zealand.

AIM: To describe the prevalence of cardiovascular disease (CVD) in New Zealand by ethnicity and socioeconomic status using NHI-linked electronic national databases. METHOD: CVD prevalence by ethnicity and socioeconomic status in New Zealand in 2006/07 were estimated from national datasets of public hospital discharges, mortality registrations, and pharmaceutical dispensing over the period 1988-2007. RESULTS: In 2007, Maori had the highest age-standardised prevalence (7.41%) compared to non-Maori, non-Pacific, and non-Indians (4.45%). Maori males and females had the highest age-specific prevalence of CVD across virtually all age groups. There was a clear gradient of increasing CVD prevalence with increasing level of social deprivation. The corresponding age-specific CVD prevalence among the least deprived quintile of Maori were similar to the most deprived quintile of 'Other' New Zealanders. CONCLUSION: Consistent with mortality trends, this study confirms marked ethnic and socioeconomic disparities in CVD prevalence that are (at least in part) independent of each other. Aggressive targeting of CVD risk management among these relatively easily identifiable high-risk patient groups with known CVD could be a highly cost-effective way of reducing health disparities in the short term.