Processes of Care and Associated Factors in Patients With Stroke by Immigration Status

Stroke is considered a leading cause of death and disability globally, with high individual, family, and societal costs.1,2 Evidence suggests that receiving guideline-recommended processes of stroke care such as early admission to stroke unit, early computed tomography/magnetic resonance imaging (CT/MR) scan, timely thrombolytic and/or endovascular therapy, early assessment and neuro-rehabilitation by physiotherapist and occupational therapist is associated with lower risk of medical complications and decreased mortality.3–6 Several studies conducted in the United States have shown disparities in receiving evidence-based stroke care processes in minorities such as African Americans relative to White Americans or Hispanic Americans.7–9 One study from Canada found no disparities in stroke care performance measures and is among the few studies up-to-date reporting data on immigrants.10 In Europe, disparities in stroke care may also be found.11,12 Hence, we have recently reported on prehospital delay, door-to-needle time, timely reperfusion therapy, and medical prophylaxis including antithrombotic, antihypertensive, and lipid-lowering therapy in immigrants with stroke in Denmark.11,12 Still, more comprehensive data on early stroke care including receiving CT/MR scan, dysphagia screening, timely mobilization, early assessment by physiotherapists, and occupational therapists among immigrants are scarce.

Disparities in the fulfilment of relevant care processes among ethnic minorities and immigrants may be attributed to patient characteristics such as socioeconomic status, language barriers, health beliefs, and acculturation factors (social, psychological, and cultural change associated with adapting into a new cultural environment), and characteristics of health delivery system such as resource and organizational factors, as previously described in Andersen behavioral model.13–18 For example, a study in the United Kingdom found that socioeconomic deprivation is associated with poor fulfilment of stroke care processes and may be stronger in patients who are Black than White.15 In Denmark, lacking language proficiency and strong cultural norms have been reported to hinder the utilization of care among Turkish and Pakistani immigrants.17,18 However, it remains largely unknown to what extent social determinants of health affect immigrants’ chances of receiving optimal stroke care.

Therefore, the present study sought to address this knowledge gap and investigated whether disparities in guideline-recommended acute stroke care exist between immigrants and Danish-born residents. Furthermore, we compared the processes of stroke care between immigrants and Danish-born residents by socioeconomic status, marital status, and duration of residence.

METHODS Study Design and Population

The Danish health care system is primarily financed through taxation, allowing free access to hospital care and general practitioners for all residents including immigrants with residence permit. This nationwide register-based cohort study was conducted using data from the Danish Stroke Registry (DSR). Reporting to the DSR is mandatory for all hospitals in Denmark providing acute stroke care. The DSR provided patient-level data on stroke type, age, sex, marital status, lifestyle risk factors, comorbidities, admission date, hospital identification number, and processes of care.19 All diagnoses were identified in the DSR in accordance with World Health Organization’s International Classification of Diseases (ICD) system, 10th revision (ICD 10). Included were patients aged 18 years or older admitted with ischemic stroke (ICD 10, I63) and hemorrhagic stroke (ICD 10, I61) starting from January 1, 2005, to December 31, 2018. Validation studies have reported a high sensitivity and positive predictive value of stroke diagnoses in the DSR.20,21

The study population was linked with registers at Statistics Denmark containing information on immigration status (country of origin), date of birth, date of immigration, highest attained education, family income, and occupation by a unique personal identification number.22–24 Only immigrants with residence permit were included. This study was approved by the Danish Data Protection Agency, reference number 514–0457/20–3000 and adhered to the principles underlying the Declaration of Helsinki. In Denmark, no further approval is required regarding registry-based research.

Immigration Status

Immigration status was constructed based on the country of origin or birth in accordance with Statistics Denmark.25 The study population was classified as immigrants or Danish-born residents. Immigrants are defined as “persons born abroad whose parents are both (or one of them if there is no available information on the other parent) foreign citizens or were both born abroad.”25 In addition, immigrants were categorized by region of origin: western and non-western immigrants. Western immigrants comprised all immigrants originating from 28 European Union countries and Andorra, Iceland, Liechtenstein, Monaco, Norway, San Marino, Switzerland, Vatican State, Australia, New Zealand, Canada, and the United States. Immigrants from other countries constituted the non-western group.25 Immigrants originating from Turkey, Poland, Iraq, and Pakistan were presented as separate subgroups because of their large proportion among the population of immigrants in Denmark. Danish-born residents were regarded as a reference group. Descendants were excluded because of small numbers.

Outcome

The process measures included in the present study were defined by the multidisciplinary national expert committee of clinical stroke experts to reflect core elements of early hospital-based care for patients with acute stroke. The following in-hospital processes of stroke care (performance measures) were included as outcomes: (1) admission to a stroke unit within 24 hours of stroke onset, (2) receipt of CT/MR scan on the day of admission, (3) dysphagia screening (direct swallowing test) on the day of admission, (4) dysphagia screening (indirect swallowing test) on the day of admission, (5) mobilization on the day of admission, (6) assessment by physiotherapist, (7) occupational therapist, and (8) nutritional risk assessment within 2 days of admission, (9) CT angiography/ultrasound within 4 days of admission, (10) administration of antiplatelets if relevant within 2 days of admission, and (11) administration of anticoagulants if relevant within 14 days of admission. Stroke care processes were further assessed using the opportunity-based composite score and the all-or-none composite score, as previously used in quality of care assessment.26 The opportunity-based composite score for each person was calculated by counting total number of care processes received (numerator), divided by total number of times that care processes were eligible for that person (denominator). The all-or-none composite score for a person was defined as 1 if all eligible care processes were received by the patient and 0 otherwise. Both composite scores were based on 11 care processes across all patients with acute stroke.

Covariates

Age, sex, hospital identification number, stroke type, stroke severity, smoking, diabetes, hypertension, atrial fibrillation, myocardial infarction, previous stroke or transient ischemic attack, marital status, education, income, occupation, and duration of residence (proxy for acculturation) were used as covariates to study association between immigration status (immigrants vs. Danish-born residents) and use of care processes among patients with stroke. Stroke severity was assessed at admission and classified as a continuous variable according to the Scandinavian Stroke Scale score. Lifestyle risk factors and comorbidities collected at baseline were categorized as yes or no. Marital status was classified as cohabiting, living alone, or other. Income was grouped into low, middle, or high tertiles. Education was categorized as low (primary and lower secondary education), medium (upper secondary, postsecondary nontertiary, and short cycle tertiary education), or high (bachelor, master, and doctoral education) according to the International Standard Classification of Education. Moreover, occupation was classified as employed, pensioner/retired, or unemployed. Duration of residence was defined as time difference between date of entry in Denmark and date of admission to a stroke unit. The duration of residence was first handled as a continuous variable in years. When evaluating overall quality of stroke care, 3 levels of duration of residence for immigrants were created: <10 years of residence, 10−20 years of residence, >20 years of residence in Denmark and a separate category for Danish-born residents.

Statistical Analyses

Processes of care among immigrants with stroke compared with Danish-born residents were summarized using frequencies and percentages. We estimated the use of 11 processes of care between immigrants and Danish-born residents by fitting multivariable logistic regression models. Models were adjusted for age, sex, clinical factors (stroke severity, previous of stroke or transient ischemic attack, smoking, myocardial infarction, atrial fibrillation, diabetes, and hypertension), and sociodemographic factors (marital status, income, occupation, education, and duration of residence). In case of missing data, we included the maximum number of patients with information for a specific covariate in the multivariable analyses. The first model contained immigration status (2 levels): overall immigrants and Danish-born residents. The second model contained region of origin (3 levels): western, non-western immigrants, and Danish-born residents. The third model consisted of country of origin: Polish, Turkish, Iraqi, Pakistani, Others, and Danish-born residents. In all models, Danish-born residents were the reference group. We reported odds ratios (ORs) and their 95% CI as unadjusted and fully adjusted.

Furthermore, overall quality of stroke care was evaluated using opportunity-based and all-or-none composite score measures. We used a linear regression when computing mean absolute differences in composite scores comparing immigrants with Danish-born residents. We assessed whether quality of care in immigrants compared with Danish-born residents differed by stroke type in a stratified linear regression. Moreover, we performed subgroup analyses in which quality of stroke care was compared between immigrants and Danish-born residents based on (1) education (4 groups: immigrants high education level, Danish-born high education level, immigrants low education level vs. Danish-born low education level), (2) income (4 groups: immigrants high income, Danish-born high income, immigrants low income vs. Danish-born low income), (3) marital status (4 groups: immigrants cohabiting, Danish-born cohabiting, immigrants living alone vs. Danish-born living alone), and (4) duration of residence (4 groups: immigrants with <10 y of residence, immigrants with 10–20 y of residence, immigrants with >20 y of residence in Denmark vs. Danish-born residents). Finally, an interaction term (hospital identifier and immigration status) was applied to assess quality of care between hospitals treating patients with acute stroke. We reported mean, mean absolute differences, and their 95% CI as percentages. Comparisons were made using Pearson χ2 test for categorical variables and Kruskal-Wallis test for continuous variables. All analyses were performed in R statistical software (version 4.0.5 and 4.1.1).

RESULTS Patient Characteristics

Between 2005 and 2018, 129,724 stroke cases were identified, of which 5796 (4.5%) were immigrants and 123,928 (95.5%) were Danish-born residents (Figure, Supplemental Digital Content 1, https://links.lww.com/MLR/C555). Compared with Danish-born residents, immigrants in particular those originating from non-western countries were in general younger at stroke diagnosis, more likely to have low income, more likely to be unemployed, and more likely to be cohabiting (Table, Supplemental Digital Content 2, https://links.lww.com/MLR/C555). Turkish were the most likely of all immigrant subgroups to have low educational level. Polish immigrants were the most likely of all immigrant subgroups to have low income. Diabetes was more prevalent in Pakistani, followed by Turkish and Iraqi than in Danish-born residents. The median Scandinavian Stroke Scale score was 49 for immigrants and Danish-born residents (Table 1).

TABLE 1 - Baseline Characteristics of Patients Admitted With Ischemic and Hemorrhagic Stroke Between 2005 and 2018 Danish-born, N=123,928 Immigrants, N=5796 Polish, N=254 Turkish, N=360 Iraqi, N=179 Pakistani, N=258 Others,* N=4745 Age at stroke diagnosis, y median (IQR) 73 (63–82) 68 (58–78) 68 (57–81) 63 (52–72) 63 (53–71) 64 (57–70) 69 (58–79) Duration of residence, y median (IQR) NA 22 (14–31) 25 (10–32) 30 (25–36) 16 (11–21) 31 (25–38) 22 (13–30) Sex (female), n (%) 66,492 (53.7) 3053 (52.7) 117 (46.1) 206 (57.2) 123 (68.7) 168 (65.1) 2439 (51.4) Ischemic stroke, n (%) 108,736 (87.7) 5039 (86.9) 227 (89.4) 324 (90.0) 157 (87.7) 232 (89.9) 4099 (86.4) Hemorrhagic stroke, n (%) 15,192 (12.3) 757 (13.1) 27 (10.6) 36 (10.0) 22 (12.3) 26 (10.1) 646 (13.6) SSS score, median (IQR) 49 (35–56) 49 (34–56) 49 (29–55) 52 (39–56) 52 (36–56) 53 (42–56) 49 (33–56)  Missing, n (%) 8496 (6.9) 546 (9.4) 27 (10.6) 31 (8.6) 20 (11.2) 38 (14.7) 430 (9.1) Comorbidities, n (%)  Current smoking 35,943 (29.0) 1545 (26.7) 67 (26.4) 112 (31.1) 47 (26.3) 61 (23.6) 1258 (26.5)  Missing 19,949 (16.1) 1057 (18.2) 55 (21.7) 54 (15.0) 27 (15.1) 42 (16.3) 879 (18.5)  Hypertension 66,493 (53.7) 3129 (54.0) 147 (57.9) 197 (54.7) 100 (55.9) 176 (68.2) 2509 (52.9)  Missing 4247 (3.4) 176 (3.0) 5 (2.0) 8 (2.2) 5 (2.8) 6 (2.3) 152 (3.2)  Diabetes 17,398 (14.0) 1366 (23.6) 51 (20.1) 140 (38.9) 57 (31.8) 153 (59.3) 965 (20.3)  Missing 3018 (2.4) 140 (2.4) 5 (2.0) 5 (1.4) 6 (3.4) 6 (2.3) 118 (2.5)  Myocardial infarction 10,558 (8.5) 507 (8.7) 17 (6.7) 39 (10.8) 18 (10.1) 46 (17.8) 387 (8.2)  Missing 4522 (3.6) 227 (3.9) 9 (3.5) 11 (3.1) 8 (4.5) 11 (4.3) 188 (3.9)  Atrial fibrillation 22,370 (18.1) 916 (15.8) 55 (21.7) 49 (13.6) 17 (9.5) 11 (4.3) 784 (16.5)  Missing 3319 (2.7) 160 (2.7) 5 (2.0) 15 (4.1) 8 (4.5) 8 (3.1) 124 (2.6)  Previous stroke or TIA 28,084 (22.7) 1228 (21.2) 59 (23.2) 66 (18.3) 34 (19.0) 59 (22.8) 1010 (21.3)  Missing 40,062 (32.3) 1723 (29.7) 60 (23.6) 128 (35.6) 58 (32.4) 63 (24.4) 1414 (29.8) Education†, n (%)  Low 52,452 (42.3) 1481 (25.5) 37 (14.6) 175 (48.6) 42 (23.5) 94 (36.5) 1133 (23.9)  Medium 13,790 (11.1) 1547 (26.7) 92 (36.2) 65 (18.1) 40 (22.3) 86 (33.3) 1264 (26.6)  High 14,614 (11.8) 880 (15.2) 40 (15.7) 9 (2.5) 35 (19.6) 22 (8.5) 774 (16.3)  Missing 43,072 (34.8) 1888 (32.6) 85 (33.5) 111 (30.8) 62 (34.6) 56 (21.7) 1574 (33.2) Family income‡, n (%)  Low 42,472 (34.2) 2512 (43.4) 140 (55.1) 164 (45.6) 72 (40.2) 95 (36.8) 2041 (43.0)  Middle 40,589 (32.8) 1781 (30.7) 58 (22.8) 106 (29.4) 73 (40.8) 94 (36.4) 1450 (30.6)  High 40,867 (33.0) 1503 (25.9) 56 (22.1) 90 (25.0) 34 (19.0) 69 (26.8) 1254 (26.4) Occupation, n (%)  Employed 33,182 (26.8) 1345 (23.2) 73 (28.7) 79 (22.0) 30 (16.8) 62 (24.0) 1101 (23.2)  Pensioner/retired 69,508 (56.1) 2686 (46.4) 112 (44.1) 148 (41.1) 51 (28.5) 100 (38.8) 2275 (47.9)  Unemployed 11,634 (9.4) 1179 (20.3) 35 (13.8) 116 (32.2) 88 (49.1) 87 (33.7) 853 (18.0)  Missing 9604 (7.7) 586 (10.1) 34 (13.4) 17 (4.7) 10 (5.6) 9 (3.5) 516 (10.9) Marital status, n (%)  Cohabiting 65,815 (53.1) 3378 (58.3) 114 (44.9) 289 (80.3) 133 (74.3) 206 (79.8) 2636 (55.6)  Living alone 50,990 (41.1) 2042 (35.2) 115 (45.3) 55 (15.3) 41 (22.9) 35 (13.6) 1796 (37.9)  Other 3904 (3.2) 179 (3.1) 13 (5.1) 7 (1.9) 5 (2.8) 8 (3.1) 148 (3.1)  Missing 3219 (2.6) 197 (3.4) 12 (4.7) 9 (2.5) NA 9 (3.5) 165 (3.4)

TIA data were available from 2013 onward.

*Other immigrant groups excluding Polish, Turkish, Iraqi, and Pakistani.

†According to the International Standard Classification of Education.

‡Tertiles.

IQR indicates interquartile range; NA, not applicable; SSS, Scandinavian Stroke Scale; TIA, transient ischemic attack.


Processes of Care

Compared with Danish-born residents, immigrants were less likely to be admitted to a stroke unit within 24 hours after stroke onset (81.5% vs. 83.9%, P<0.001), with the lowest proportion seen in Iraqi. Immigrants less frequently received stroke care than Danish-born residents (70.6% vs. 73.8%, P<0.001 for early direct swallowing test, 81.3% vs. 83.3%, P=0.003 for early indirect swallowing test, 79.2% vs. 81.3%, P=0.001 for early physiotherapy, 77.6% vs. 79.4%, P=0.004 for early occupational therapy, 82.0% vs. 83.5%, P=0.011 for early nutritional assessment; Table 2) (Table, Supplemental Digital Content 3, https://links.lww.com/MLR/C555). The lowest proportions of receiving early dysphagia screening (direct or indirect swallowing tests) and early physiotherapy and occupational therapy were found in Pakistani, Polish, and Turkish immigrants (Table 3). Compared with Danish-born residents, we observed lower unadjusted odds of receiving early dysphagia screening (OR: 0.97; 95% CI, 0.96–0.98), early physiotherapy (OR: 0.97; 95% CI, 0.96–0.99), early occupational therapy (OR: 0.98; 95% CI, 0.97–0.99), and early nutritional assessment (OR: 0.98; 95% CI, 0.97–0.99) among immigrants (Fig. 1) (Table, Supplemental Digital Content 4, https://links.lww.com/MLR/C555). After adjustment for age, sex, clinical, and sociodemographic factors, the odds ratio of early stroke unit admission (OR: 0.97; 95% CI, 0.94–0.99), early dysphagia screening (OR: 0.96; 95% CI, 0.93–0.98), early physiotherapy (OR: 0.96; 95% CI, 0.94–0.99), and early occupational therapy (OR: 0.96; 95% CI, 0.93–0.98) were significantly lower for immigrants than Danish-born residents (Fig. 1) (Table, Supplemental Digital Content 5, https://links.lww.com/MLR/C555). The lower odds of early dysphagia screening, early physiotherapy, and occupational therapy were most evident in non-western immigrants. In the country-specific estimates, the lower odds of receiving the aforementioned stroke care processes were more pronounced for Turkish, Iraqi, and Pakistani immigrants than Danish-born residents. In contrast, immigrants were more likely to receive CT angiography/ultrasound and had a higher odds of receiving this care process than Danish-born residents in the unadjusted model (84.3% vs. 82.0%, P<0.001 and OR: 1.02; 95% CI, 1.01–1.03); however, the association disappeared after adjustment for covariates. No evidence was found that other stroke care processes differed between immigrants and Danish-born residents.

TABLE 2 - Processes of Stroke Care by Immigration Status Total Danish-born Immigrants Processes of care Eligible, N Eligible, N Received care, N (%) Eligible, N Received care, N (%) P Admission to a stroke unit within 24 h 129,724 123,928 103,941 (83.9) 5796 4724 (81.5) <0.001 CT/MR scan on the day of admission 127,211 121,547 97,250 (80.0) 5664 4542 (80.2) 0.739 Dysphagia screening on the day of admission (direct swallowing test) 86,528 82,643 61,004 (73.8) 3885 2741 (70.6) <0.001 Dysphagia screening on the day of admission (indirect swallowing test) 65,516 62,418 52,022 (83.3) 3098 2519 (81.3) 0.003 Mobilization on the day of admission 67,619 64,336 50,258 (78.1) 3283 2531 (77.1) 0.166 Physiotherapy within 2 d of admission 101,967 97,660 79,371 (81.3) 4307 3410 (79.2) 0.001 Occupational therapy within 2 d of admission 103,148 98,775 78,424 (79.4) 4373 3393 (77.6) 0.004 Nutritional risk assessment within 2 d of admission 101,939 97,473 81,358 (83.5) 4466 3663 (82.0) 0.011 CT angiography/ultrasound within 4 d of admission 73,102 69,596 57,062 (82.0) 3506 2956 (84.3) <0.001 Antiplatelet therapy within 2 d of admission 85,310 81,472 74,272 (91.2) 3838 3522 (91.8) 0.197 Anticoagulant therapy within 14 d of admission 12,446 11,952 10,118 (84.7) 494 409 (82.8) 0.261 Opportunity-based composite score, reported as percentage, mean (95% CI) 129,724 123,928 79.9 (79.8–80.1) 5796 78.9 (78.2–79.6) <0.001 All-or-none score, reported as percentage, mean (95% CI) 129,724 123,928 45.7 (45.5–46.0) 5796 44.0 (42.7–45.3) <0.001
TABLE 3 - Processes of Stroke Care by Country of Origin Danish-born Polish Turkish Iraqi Pakistani Others* Processes of care Eligible, N Received care, N (%) Eligible, N Received care, N (%) Eligible, N Received care, N (%) Eligible, N Received care, N (%) Eligible, N Received care, N (%) Eligible, N Received care, N (%) P Admission to a stroke unit within 24 h 123,928 103,941 (83.9) 254 203 (79.9) 360 298 (82.8) 179 138 (77.1) 258 206 (79.8) 4745 3879 (81.7)

留言 (0)

沒有登入
gif