a b s t r a c t

Objective: To evaluate whether a fast-track intervention program will reduce time-lags of patients with STEMI considering minority groups, various socioeconomic status (SES) and clinical risk factors.

Methods: A retrospective-archive study was conducted according to clinical guidelines, comparing all STEMI pa- tients (n = 140) admitted to the emergency department (ED) before (n = 60) and during (n = 80) implemen- tation of the fast track intervention program. The program comprised four steps: (1) immediate bed rest,

(2) marking patient chart, (3) assessing time-lags according to defined clinical guidelines, and (4) physician signing a dedicated sticker on the ECG. Results: The major ethnic group compared to other minority patients with STEMI were less delayed for physician examination (r = -0.398, p b 0.01), spent less time at ED (r = -0.541, p b 0.01) and reached percutaneous cor- onary intervention earlier (r = -0.672, p b 0.01). Patients with higher SES spent less time for physician (r =

-338, p b 0.05) and in the ED (r = -0.415, p b 0.01). Before intervention patients with diabetes mellitus (DM) spent more time at ED compared to non DM patients, however during intervention this difference was blurred (? = -0.803, p b 0.001). Gaps regarding sociodemographic bias remained present throughout the inter- vention despite monthly staff evaluations considering patient cases.

Conclusions: The fast track intervention was associated with less time at ED and to cardiac reperfusion. Yet, sociodemographic bias was present. Our findings highlight the need for the healthcare profession to address the role of biases in disparities in healthcare.

(C) 2019

Background

Time delay from the onset of cardiac symptoms to reperfusion in pa- tients presenting with ST elevation myocardial infarction is a major poor prognostic factor [1-6]. Globally, delayed diagnosis occurs in 5% to 15% of patients diagnosed with STEMI [7]. STEMI is defined as a class I indication for Door-to-balloon time (DTBT) to initiate percuta- neous coronary intervention (PCI) within 90 min [5,8-10]. Customarily, patients admitted to the emergency departments (EDs) with chest pain should undergo a rapid triage assessment and treated under high- priority scoring [11-14]. The American Heart Association (AHA) and

? The research has not been presented. All authors attest to meeting the ICMJE.org authorship criteria.

* Corresponding author at: Rambam Health Care Campus & School of Nursing, Haifa University, Israel.

E-mail address: [email protected] (M. Saban).

the American College of Cardiology (ACC) recommended time-lag guidelines for patients presenting in the ED with symptoms suggestive of STEMI. Guidelines include obtaining an electrocardiogram (ECG) within 10 min, evaluating the patient by medical staff within 15 min, and receiving troponin blood test results within 60 min from arrival [15-17]. However, in some cases, these patients are postponed, receiv- ing lower priority score, delayed from STEMI diagnosis and timely PCI treatment, and therefore may suffer from Prolonged hospitalization and exceeded mortality rates [3,6,7,13,18].

Several triage classification tools are used in EDs to determine the patient urgency [13,18]. According to the Canadian Triage and Acuity Scale (CTAS), patients classified as P1 require immediate treatment, and patients classified as P2 to P5 are expected to receive medical as- sessment and treatment within 15, 30, 60, and 120 min, respectively [13]. Ultimately, patients with STEMI should be classified as P1 or P2. A few studies have further shown a significant reduction in DTBT for pa- tients correctly classified by Triage nurses‘ [13,14]. Yet, these studies did

https://doi.org/10.1016/j.ajem.2019.158380

0735-6757/(C) 2019

not assess the whole triage process, which include, beyond patient’s ur- gency classification, time-lags to ECG, to physician, and to troponin blood test results.

DTBT is associated with ED setting factors including workload, shift type (morning, evening, or night), and whether the patient attended during shift handover [19-24]. Overcrowded EDs have been thoroughly described and studied [25-27]. This issue is universal and most EDs ex- ceed their planned maximal capacity. [11,26,27] It may leads to error in triage classification, causing prolonged ED Waiting times, and adversely impacting DTBT [11,13,14,26,27]. There are several strategies for de- creasing DTBT, such as activating the catheterization lab in the cardiol- ogy department, by an ED physician without consulting a cardiologist [7].

Despite the known importance of fast diagnosis and the diversity of intervention programs in STEMI patients, their impact on patient char- acteristics, SES, affiliation to minority groups and clinical risk factors re- mains unclear.

Multiple factors may potentially mediate this disparity, including a higher prevalence of Cardiovascular risk factors, inequalities in access to cardiac investigations, including angiography and PCI and poorer compliance with medical therapy in patients with socioeconomic disad- vantage [6,28]. However, the association between minority groups SES, and STEMI therapy is less well understood due to a paucity of evidence and conflicting data. Thus, given the inconsistencies in the literature and an increasing focus worldwide to reduce inequalities in health out- comes, we sought to evaluate the impact of a fast-track intervention program on reducing time-lags in patients with STEMI considering mi- nority groups, various SES and clinical risk factors.

Methods

Study design and setting

A retrospective study was conducted from January 2015 to Decem- ber 2016, in the ED of a tertiary hospital in the country of Israel, after ethical approval. The ED consists of 100 beds and serves about 130,000 patients over 18 years old on average per year. Of these, about 5500 (5%) patients arrive with annual chest pain and about 80 (1.5%) are diagnosed with STEMI [29].

Participants

The study sample consisted of 140 patient files who were triaged and diagnosed as a possible STEMI at the ED during 2015 (n = 60) and 2016 (n = 80).

Most of the patients present via ambulance (77%) while others were walk-in.

Files were extracted from the electronic database by using the ICD- 10 code for STEMI in ED transfer as a filter (i21.3, “acute transmural myocardial infarction of unspecified site”). Overall sample was drawn from a total of 335 patient files (170 in 2015 and 165 in 2016) who were hospitalized with STEMI. Patients who were transferred directly to the intensive coronary care unit were excluded from the study (n

= 186).

Intervention and procedure

The fast-track intervention program for patients with chest pain was implemented between January and December 2016. The fast-track composed of two main domains. First, a set of clinical guidelines, ac- cording to the American Nurses Association (ANA) and ACC [8], for pa- tients with chest pain admitted to an ED was tailored and implemented to include 15? to nursing triage, then 10? to ECG, 40? to physician assess- ment, 60? waiting time to decision, and 90? to DTBT [8,13,17]. In addi- tion, adjustments were made in the electronic medical records including, automatic notification and time exceeding alerts. Second,

the core program comprised of four steps: (1) immediate bed rest,

(2) marking the patient chart with a dedicated sticker (Appendix 1),

(3) assessing the time-lags according to defined clinical guidelines, and (4) signing a dedicated sticker on the ECG by the physician (Appendix 2).

The fast-track intervention was conducted after a training session of the ED staff during three rounds of nursing-physician staff meetings. The staff was introduced to the program aim and procedures; were trained on the time-lag guidelines; and on the four steps of the program. During the first month of the intervention, a brief training was held and a short message service (SMS) reminder was disseminated before each shift, by the first author. In addition, each month, cases that upheld the clinical guidelines were distributed to hospital staff by internal mail list. An anonymous list of STEMI patients was sent via an internal mail net- work on a monthly basis for peer evaluation and knowledge. Attached to each mail was a descriptive table containing patient characteristics including age, gender, ethnicity, day of the week and shift type. Cases that failed to meet the criteria underwent a full inquiry by the hospital Safety and Quality Committee, which includes the involved ED staff and the cardiologic team. During the inquiry, each step of the patient’s treatment was re-evaluated and studied to guide further cases. Yet, as each case was presented and addressed individually the staff was not able to indicate a specific variation regarding this matter.

All data were collected retrospectively between January and March 2017, from electronic medical records, by the first author. For each pa- tient, we collected and measured the following: Outcome variables, namely total time in ED, waiting time between the ED exam and referral to cardiac catheterization (GAP), DTBT, LOS in hospital, and mortality rates.

Based on the main correlation [30] and the current literature [6,11,13,31], Independent variables consisted of Assessment characteris- tics, composed of P-scale classification and time to nursing triage, to ECG, to physician; ED setting variables, including ED workload (a dichot- omous variable, where 1 = over 300 patients, and 0 = b300 patients treated in a single day); day of the week, shift type (morning, evening, and night), whether the patient attended during shift handover; and Clinical risk factors for STEMI, including smoking, dyslipidemia, hyper- tension (HTN), diabetes mellitus (DM), previous cardiac events and family history of coronary heart disease . Sociodemographic char- acteristics including gender, age, SES and ethnicity descent (Jewish vs. minorities).

The weighted census population included 8,985,215 residents living in Israel in 2019, of which 51% are females, 28.3% are children under age 15 and 9.6% are elderly ages 65+. The non-Jewish minority comprises 24.6% of the population. Of them, 21% are Arabs, while 3.6% defined as ‘others’. In the current study both of these groups were referred to as minorities. The SES of Arabs and other communities is lower than that of the majority of Jewish communities [32].

Data management and statistical analysis

All continuous guideline variables and DTBT were cut off according to the abovementioned defined time guidelines to construct dichoto- mized variables which adhere to the ANA and AHA recommendation. These variables were also analyzed continuously. Associations between total time in ED, time waiting between the STEMI diagnosis at ED and referral to cardiac catheterization, and DTBT, and each of the study inde- pendent ED Assessment characteristics, ED setting variables, Clinical risk factors and sociodemographic characteristics, were examined before and during the intervention implementation, using chi-square tests for cat- egorical variables and t-tests or one-way analysis of variance (ANOVA), when appropriate, for continuous variables (Table 1). Pearson’s coeffi- cient was used to calculate statistical correlations between the variables employed in the study either ordinal or dummy before and during implementing the intervention program (Tables 2A and 2B). Addition- ally, sub-group analysis was performed to evaluate time per each part

Table 1

Sociodemographic characteristics, Clinical risk factors, ED setting and assessment charac- teristics and outcomes before and during implementing the fast-track program for pa- tients with STEMI

(e.g., time waiting in the ED, time waiting between the STEMI diagnosis at ED and referral to cardiac catheterization-GAP, and DTBT), we con- ducted three linear regression analyses in three steps (Table 4). First

the intervention variable was entered into the model. In the second

Characteristics Before

intervention (2015) n = 60

Sociodemographic Age (mean +- SD) Gender (n, %)	63.12 +- 13.16	63.99 + 13.44	0.89 cardiac events and priority score (P) according to CTAS) were entered into the model. Leading Organizational characteristics that influence
Male	26 (56.7%)	35 (43.8%)	0.96 the triage decision making, such as workload and shift, were not in-
Female Ethnicity (n, %)	34 (43.3%)	45 (56.2%)	cluded in the model due to their confounding potential [13,33]. In the third step, the interactions between the intervention and each

After intervention (2016) n = 80

P-value

step, the Sociodemographic variables (i.e., age, minority group, and SES) and the clinical risk factors (i.e., smoking, dyslipidemia, high blood pressure, diabetes, family history of myocardial infarction, prior

C

Jews	16 (26.7%)	27 (33.7%)
Arabs	39 (65%)	48 (60%)	0.63
Others	5 (8.3%)	5 (6.3%)
Socio Economic Status (median 7.00 + IQR) (6.00-7.00)		7.00 (6.00-7.00)	0.16
linical risk factors Smoking (n, %) 31 (51.7%) 33 (41.3%) 0.44
Dyslipidemia (n, %)	33 (55%)	40 (50%)	0.52
Hypertension (n, %)	27 (45%)	49 (61.3%)	0.06
Diabetes mellitus (n, %)	18 (30%)	28 (35%)	0.55
FH of CHD (n, %)	23 (38.3%)	26 (32.5%)	0.48
No. of cardiac events (mean +- SD)	1.34 +- 0.66	1.43 +- 0.84	0.20

ED setting

Workload (n, %) D

N300 patients at ED/day 53 (88.3%)		69 (86.3%)	0.72
ay in the week (n, %) Sunday 13 (21.7%) 13 (16.3%)
Monday to Thursday	33 (55%)	42 (52.5%)	0.51
Friday and Saturday	14 (23.3)	25 (31.3%)

Type of shift (n, %)

Morning (7 am-3 pm)	25 (41.7%)	32 (40%)	fast-track intervention program. About 50% of the participants were
Evening (3 pm-11 pm)	20 (33.3%)	27 (33.8%)	0.97 male (56.7% in 2015 and 43.8% in 2016), and approximately 60% were

of the variables in the second step were entered into the model. For each regression model, we calculated the explained variance and the F value for the variance. The level of significance for all statistical analyses was 5%. We performed the data analysis using the Statistical Package for Health & Welfare Science for Windows (SPSS, version 25.0, Chicago, IL, USA).

Results

Baseline characteristics

Table 1 shows the sociodemographic characteristics, clinical risk fac- tors, ED setting, and ED assessment characteristics of the participants, be- fore and during the fast-track intervention program. No statistical differences were found in any of the sociodemographic characteristics, clinical risk factors or the ED setting measures, before and during the

Night (11 pm-7 am) 15 (25%) 21 (26.3%)

Arrived in shift handover (n, %) 1 (1.7%) 4 (5%) 0.29

ED assessment

P-scale (n, %)

1	6 (10%)	9 (11.3%)
2	12 (20%)	31 (38.8%)
3	39 (65%)	31 (38.8%)	0.02
4	3 (5%)	6 (7.5%)
5	0	3 (3.8%)
P1-2	18 (30%)	40 (50%)
P3-5	42 (70%)	40 (50%)	0.01

of Arab minority group. SES indices were similar during the study pe- riods (median = 7.00; IQR 6.00-7.00). About 40% of the STEMI patients arrived during mornings’ shifts (41.7% and 40% in 2015 and 2016, re- spectively) and only a few were admitted during shifts’ handover (1.7% and 5%, respectively). About half of the patients arrived in the middle of the week, in both 2015 and 2016.

Post interventional characteristics

Time to triage (n, %) <=15? N15?	43 (71.7%) 17 (28.3%)	64 (80%) 16 (20%)	0.25 In 2015, before the intervention, only 30% (n = 18) of patients were triaged according to the P-scale guidelines, namely P <= 2, whereas dur-
Mean +- SD	14.22 +- 12.79	10.21 +- 7.98	0.03 ing the intervention, in 2016, 50% (n = 40) of patients were classified

Time to ECG (n, %)

<=10?	24 (40%)	46 (57.5%)	0.04
N10?	36 (60%)	34 (42.5%)
Mean +- SD	18.48 +- 14.31	11.96 +- 8.93	b0.001
ime to physician (n, %) <=40? 49 (81.7%) 66 (82.5%) 0.89
N40?	11 (18.3%)	14 (17.5%)
Mean +- SD	33.75 +- 28.35	24.79 +- 16.92	0.03

T

Outcomes

Total time in ED (n, %)

according to the correct P-scale category. This difference was statisti- cally significant (p = 0.01). While before the intervention only 40% of the patients underwent an ECG examination within 10 min, and 63.3% were held up in the ED for a decision b60 min; during the intervention- the rates improved to 57.5% (p = 0.04) and 87.5% (p = 0.01), respectively. The fast-track program also decreased the mean time-lags for nurse triage (14.22 +- 12.79 in 2015 to 10.21 +- 7.98 in

2016, p = 0.03); for physician examination (33.75 +- 28.35 in 2015 to

<=60?	38 (63.3%)	70 (87.5%)	b0.001	24.79 +- 16.92 in 2016, p = 0.03), for total time in ED (71.95 +- 56.38
N60?	22 (36.7%)	10 (12.5%)		in 2015 to 37.40 +- 19.62 in 2016, p b 0.001), and for DTBT (106.32 +-

60.46 in 2015 to 79.90 +- 38.13 in 2016, p b 0.001). No statistical differ- ences were found in LOS after PCI or in mortality rates before and during

Mean +- SD	71.95 +- 56.38	37.40 +- 19.62	b0.001
GAP (Mean +- SD)	37.87 +- 26.53	42.50 +- 30.72	0.36
DTBT (n, %)

<=90?	37 (61.7%)	56 (70%)	0.30 program intervention. However, Poisson regression indicted that adher-
N90?	23 (38.3%)	24 (30%)	ence to DTBT predict less LOS after the implementation the program in-

Mean +- SD	106.32 +- 60.46	79.90 +- 38.13	b0.001
LOS in hospital (Mean +- SD)	5.43 +- 3.16	5.89 +- 3.20	0.73
Mortality rates (n, %)	8 (13.3%)	6 (7.5%)	0.25

FH–Family History; CHD–Coronary Heart Disease; LOS–Length of Stay. GAP- time waiting between the ED exam and referral to cardiac catheterization.

of the protocol for males of majority versus those in the minority group (Table 3). Furthermore, to determine the effect of the intervention, the independent variables, and their interactions with the intervention on each of the three waiting periods that STEMI patients contend with

tervention (B = -1.080; p = 0.021).

Sub group and multivariate analysis

Tables 2A and 2B demonstrates the correlations between the vari- ables examined in the study. Findings presented describe correlations examined in 2015 (Table 2A) and the correlations examined in 2016 (Table 2B). In 2015, Jewish compared to other minorities patients with STEMI, were younger (r = -0.327, p b 0.05), less smoking (r =

-0.366, p b 0.01) and diagnosed with HTN (r = 0.523, p b 0.01).

Table 2A

Correlations between the variables employed in the study period (intervention phase).

Variables	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15 16
Sociodemographic
1. Age	–
2. Jewish	-.327?	–
3. Socio Economic Status	.096	.094	–
Clinical risk factors
4. Smoking	-.366??	.191	-.125	–
5. Dyslipidemia	.178	-.25	-.273?	-053	–
6. HTN	.523??	-.125	-.027	-.199	.344??	–
7. DM	.257	-.074	.019	-.219	.283?	.345??	–
8. FH	-.175	-.218	.095	.004	.041	-.145	.013	–
9. Prior cardiac events	.172	.228	-.070	.160	.245	.194	.158	-.194	–
10. P scale	.011	-.224	.016	.790	-.175	-.039	.011	.191	-.183	–
Time for treatment 11. Time to triage	.125	-.265	-.041	-.089	.065	.278?	.269??	.240	-.060	-.066	–
12. Time to ECG	.083	-.299?	-.140	-.190	.093	.310?	.332?	-.122	-.155	-.044	.841?	–
13. Time to physician	.004	-.247	-.185	-.025	.102	.194	.212	-.067	-.125	.142	.313?	.276?	–
14. Total time in ED	.126	-.253	.028	.015	.166	.219	.381??	-.093	-.097	.160	.294?	.378??	.687??	–
15. GAP	-.198	-.195	-.181	-.096	-.402	.023	-.125	-.086	-.408	.216	-.018	-.006	.145	.035	–
16. door to balloon time	-.098	-.285?	-.092	-.058	.106	.175	.288?	-.109	-.104	.225	.313?	.415??	.695??	.899??	.470?? –

*GAP- Time waiting between the ED exam and referral to cardiac catheterization Data represents the Pearson or Spearman.

*pb0.05

**pb0.01

***pb0.001

Furthermore, Jewish patients reached ECG examination rapidly (r =

-0.299, p b 0.05) and arrived earlier to the cardiac catheterization lab (DTBT) (r = -0.285, p b 0.05) than patients of minorities groups. In ad- dition, patients with HTN reached the nurse triage later (r = 0.278,p b 0.05) as well as ECG examination (r = 0.310, p b 0.05). Likely, patients with DM also reached the nurse triage and ECG examination later (r = 0.269, p b 0.01; r = 0.332, p b 0.05, patients with STEMI were less de- layed for physician examination (r = -0.398, p b 0.01), spent less time at ED (r = -0.541, p b 0.01) and most important reached the car- diac catheterization lab (DTBT) earlier (r = -0.672, p b 0.01). Further- more, patients with higher SES spent less time for physician (r = -338, p b 0.05) and in the ED (r = -0.415, p b 0.01). Interestingly, during in- tervention, patients with FH of CAD reached the cardiac catheterization lab (DTBT) earlier (r = -0.241, p b 0.05). As can be seen in Table 3,

significant differences were found in all time lags among males of ma- jority versus those of minority group, except in time to triage. Table 4 describes three regression models aimed at exploring factors that pre- dict the waiting period in the ED, time lapse between leaving the ED and reaching the cardiac catheterization lab, and DTBT. As for waiting time in the ED (Table 4 Model 1), in the first step, findings demonstrated that the intervention had a significant effect (? = -0.371, p b 0.001), and explained 13.8% of the variance of the ED waiting time. In the sec- ond step, risk factors were entered into the model, all of which added to the explained variance 13.8% (in total r² = 27.6%). In this step, Jewish patients and males predicted less total time at ED (? = -0.376, p b 0.001; ? = -0.241, p b 0.001; respectively), and patients diagnosed with HTN were waiting a significantly longer period of time at ED (?

= 0.180, p b 0.05). In the third step, interactions between risk factors

Table 2B

Correlations between the variables employed in the study period (post-intervention phase).

Variables	1	2	3	4	5	6	7	8	9	10	11	12	13	14 15 16
Sociodemographic 1. Age	–
2. Jewish	-.114	–
3. Socio Economic Status	.041	.272*	–
Clinical risk factors 4. Smoking	-.390**	.178	.148	–
5. Dyslipidemia	.059	-.058	-.102	-.085	–
6. HTN	.340**	-.110	.031	-.104	.220*	–
7. DM	.310**	-.087	-.116	-.062	.248*	.189	–
8. FH	-.349**	.118	.126	.256*	-.088	-.143	-246*	–
9. Prior cardiac events	.004	.081	-.038	-.024	.374**	.266**	.184	.155	–
10. P scale	-.023	-.134	-.179	-.007	-.070	-.012	.094	.030	-.148	–
Time for treatment
11. Time to triage	-.096	.041	-.151	-.038	.104	.043	.251*	-.029	.194	-.058	–
12. Time to ECG	.124	-.182	-.269*	-.088	.073	.050	.118	-.027	.113	.114	.680**	–
13. Time to physician	.190	-.398**	-.338**	-.291**	.202	.187	.137	-.147	.111	.260*	.324**	.604**
14. Total time in ED	.131	-.541**	-.415**	-.269*	.172	.161	.106	-.131	.132	.213	.164	.486**	.781**	–
15. GAP	.164	-.492**	-.063	-.044	-.116	.129	-.010	-.216	-.166	.059	-.201	-.047	.093	.104 –
16. Door to balloon time	.200	-.672**	-.163	-.174	-.004	.187	.046	-.241*	-.065	.157	-.077	.208	.477**	.598** .859** –

*GAP- Time waiting between the ED exam and referral to cardiac catheterization Data represents the Pearson or Spearman.

*pb0.05

**pb0.01

***pb0.001

Table 3

Differences in time lags among males of majority versus those of minority group

As for DTBT (Table 4 model 3), all in all, the model explained 35.8% of variance. Of this, the intervention contributed 7% and significantly re-

Time lags Majority

(n = 19)

Time to triage (n, %)

<=15?	19	35	0.062
N15?	0	7
Mean +- SD	9.16 +- 5.31	12.10 +- 11.5	0.294

Time to ECG (n, %)

Minority (n = 42)

P-value

duced the total waiting time for DTB (? = -0.265, p b 0.001). The sec- ond step demonstrated that patients of Jewish decent reached the cardiac catheterization lab earlier (? = -0.420, p b 0.05). Step 3 shows better outcomes only for patients with DM. For these patients the intervention reduced the waiting time for DTB (? = -0.600, p b 0.05).

<=10? 13 17 0.040

N10? 6 25

Mean +- SD 9.21 +- 5.70 14.71 +- 11.5 0.048

Time to physician (n, %)

<=40?	19	35	0.062
N40?	0	7
Mean +- SD	14.79 +- 9.64	28.40 +- 19.7	b0.001
Total time in ED (n, %) <=60?	19	33	0.026
N60?	0	9
Mean +- SD DTBT (n, %)	27.4 +- 10.13	54.1 +- 31.8	b0.001

Discussion

This study aimed to assess the impact of a fast-track program inter- vention on the clinical outcome of patient with STEMI considering mi- nority groups, various socio demographic and clinical risk factors. Our intervention program was based on accumulated evidence emphasizing the urgency for rapid assessment, with defined time limits of patients with suspected STEMI [8,12,15,34].

From one point of view, the results revealed a significant improve-

<=90?	18	29	0.024	ment of critical indicators for quality of care, including DTBT and ED
N90?	1	13		time-lags. This improvement is meaningful and adds to a series of stud-
Mean +- SD	51.2 +- 18.6	41.1 +- 34.4	0.047	ies that investigated strategies to improve DTBT [35], and correlates

GAP- time waiting between the ED exam and referral to cardiac catheterization; DTBT- Door to balloon time.

and the intervention were entered into the model. This step added 7% for the explained variance in waiting time at ED, and the model as a whole explained 34.6% of variance. A significant effect of the two-way interaction between the intervention and patients with DM was found, suggesting that whereas before the intervention patients with DM spent longer time in the ED, during the intervention these patients’ waiting time in ED was significantly reduced (? = -0.803, p b 0.001; Fig. 1).

As for time lapse between leaving the ED and starting the cardiac catheterization procedure-GAP (Table 4, Model 2), we found significant two-way interactions between the intervention and minority groups, and between the intervention and SES (? = 0.936, p b 0.05 and ? = 0.913, p b 0.05, respectively). Fig. 2 show that although weak, the inter- vention adversely impact patients of minority groups and of lower SES.

leading countries’ reports in this category, which showed that interven- tion in the ED is crucial to coronary reperfusion catheterization. More specifically, our results demonstrated that accurate, timely P-scale clas- sification in patients suspected of STEMI and ECG has a significant im- pact on DTBT. These findings are expected given the urgency of diagnosing and treating patients attending with STEMI [6,8,36]. Never- theless, half the patients were not classified correctly. This finding is somewhat weak, yet it may reflect a ‘trial period’.

On the other hand, our result shows that there are three groups that did not benefit from the intervention, and even endured prolongation in time lags. The first group are minorities, demonstrating prolongation in time for ECG as well as DTBT. These findings align with prior research on ethnic disparity and evidence of increasing recognition of ethnic bias in other aspects of medical management, including the ED setting, which adds external validity to these findings [37-40]. The Institute of Medi- cine report, “Unequal Treatment”, concluded that unrecognized bias against members of a social group, such as racial or Ethnic minorities,

Table 4

Linear regression models that predict the waiting period in the ED, time lapse between leaving the ED and reaching the cardiac catheterization lab (GAP), and DTBT.

Model 1 – EDT Model 2 – GAP Model 3 – DTBT

	Variable	B	SE B	?	B	SE B	?	B	SE B	?
Step 1	Intervention	-30.948	6.923	-.371***	2.679	5.343	.045	-27.82	9.089	-.265***
	R2	.138			.002			.070
	F for change in R2	19.98***			.251			9.35***
	?R2	.138			.002			.070
Step 2	Ethnicity	-31.347	6.738	-.376***	-25.293	5.380	-.407	-46.001	8.823	-.420***
	Gender (1male/0female)	-21.067	7.193	-.241***	-5.706	8.658	-.097	-17.003	14.198	-.164
	Socio Economic Status	-13.042	11.146	-.158	.707	1.624	.038	-.305	2.663	-.009
	FH (yes/no)	-.827	2.153	-.032	-6.067	9.144	-.100	-.504	14.995	-.005
	Smoking (yes/no)	7.314	11.822	.086	1.577	5.302	.026	1.646	8.695	.015
	Dyslipidemia (Yes/no)	.075	7.120	.001	-6.625	5.248	-.113	-2.929	8.607	-.028
	DM (yes/no)	3.556	7.043	.043	-4.775	5.470	-.077	10.972	8.970	.101
	HTN (yes/no)	15.621	7.342	.180*	4.957	5.290	.084	10.618	8.675	.102
	R2	.276			.211			.272
	F for change in R2	2.78***			3.85***			1.12
	?R2	.138			.209			.061
Model 3	Ethnicity*intervention	-10.379	14.631	-.281	24.583	11.034	.936*	14.424	18.257	.312
	Gender*intervention	-19.172	22.638	-.529	1.823	18.373	.071	-14.164	30.401	-.312
	Socio Economic Status *Intervention	-5.681	4.355	-.548	6.732	3.328	.913*	1.311	5.507	.101
	FH (yes/no)*intervention	-3.883	23.648	-.076	-3.953	19.132	-.108	-4.586	31.657	-.071
	Smoking (yes/no)*intervention	.350	4.354	.007	-.467	3.292	-.013	-.244	5.447	-.004
	Dyslipidemia (Yes/no)*intervention	-3.547	14.628	-.074	-6.961	11.022	-.204	-10.462	18.238	-.174
	DM (yes/no)*intervention	-40.236	15.023	-.803***	2.909	11.352	.082	-37.727	18.783	-.600*
	HTN (yes/no)*intervention	2.588	14.482	.056	-2.317	10.911	-.071	.273	18.054	.005
	R2	.346			.272			.358
F for change in R2		1.47	1.12			.861***
?R2		.007	.061			.004

120

100

80

60

40

20

0

Yes No

No intervention 98.67

55.4

Intervention 40.82

36.56

Yes No

Fig. 1. interaction between DM and time in ED, before and during the intervention.

may affect communication or the care offered to those individuals [41]. In spite of the fact that staff members were updated on patient’s cases and characteristics, alterations in treatment remained. Thus, the poten- tial damaging effect bias has on providing a gold standard of care for all patients rises. In the past decade, several interventions to reduce bias in health care team were published [41-43]. Main research found that in order to raise awareness of the potential conflict between holding neg- ative explicit attitudes towards some patient characteristics, clinicians must be thoughtful, consider others perspectives and work together to achieve common goals [42]. Yet, in the hectic and demanding medical practice it may not be so simple. With the intent to minimize bias, con- tinuous quality improvement literature emphasized teamwork as a crit- ical determinant [44]. A true collaborative process is characterized by design planning, learning and is a context in which the problem is de- fined during the process. [45] This makes all healthcare members re- sponsible for improvements in an ongoing manner.

Schrader et al. [40] found that minority patients are classified with low priority score in triage setting. Another study examining the role of racial differences among 598,911 patients hospitalized with myocar- dial infarction between 1994 and 2002 found that the rate of reperfu- sion therapy is lower in minority patients [46]. Thus, despite addressing a life threatening condition as STEMI for which there are

clear guidelines available that should allow equal and effective treat- ment, minority and low SES groups demonstrated alterations in time lags.

In addition, male patients presenting with STEMI predicted de- creased total time at ED thus increasing their likelihood to meet DTBT. gender differences found in this research are similar to findings of sev- eral studies [47]. As such, Arslanian-Engoren et al. [48] identified gender as a predictor of accurate decision making, with men more likely to be assigned an accurate designation than women despite presenting iden- tical symptoms in a vignette. The intervention adversely impact patients of lower SES. Previous studies sought to identify the impact of specific socioeconomic factors on clinical treatment [49,50]. For example, study that explored the subsequent STEMI management and outcomes in STEMI patients undergoing PCI found that symptom to-first-medical- contact time is longer among people with a lower SES [49].

The use of assessment measures guidelines, such as DTBT, represents health care systems‘ efforts to standardize the management of care to decrease morbidity and mortality-related systems risk factors, espe- cially in life-threatening conditions [10,12,35,36]. However, this ap- proach can concomitantly limit the weight of clinical risk factors in assessing and scoring patients’ urgency. The fast-track intervention re- flects this gap and resulted in less contribution for the clinical risk factors

	No intervention			Intervention
Low	37.69			42.54
High	40.25			39
		Low	High

Fig. 2. Interaction between socio economic status and time lapse between leaving the ED and starting PCI, before and during the intervention.

43

42

41

40

39

38

37

36

35

(C-statistics of 0.687 in 2016 vs. 0.732 in 2015) and increased contribu- tion for the ED settings factors category (C-statistics of 0.791 in 2016 vs. 0.783 in 2015).

Paradoxically, although not significant, during implementation of the fast-track program, more patients with STEMI were diagnosed dur- ing morning shifts than during evening and night shifts, and fewer pa- tients exceeded the 90-minute guideline. These results corresponded with recent findings indicating flawed quality of care during evening and night shifts [51]. This is particularly noteworthy since these shifts have the lowest patient workload in the ED, including in the current study [13,25,26]. Unexpectedly, our results showed that workload is positively linked to on-time DTB. This result contradicts the accepted lit- erature so far, which holds that workload causes negative health out- comes [25-27]. This finding may also be related to the availability of the catheterization team during the morning shift.

Altogether, adherence to on-time DTB improved from 61.7% in 2015 to 70% in 2016. These results are below the 85% accepted rates for DTBT guidelines but reflect the exclusion of patients with STEMI who directly attended the PCI lab. The rates are expected to be higher when combin- ing patients diagnosed with STEMI who were attending the ED with those who arrived at the PCI lab directly, as most studies reported [35,37].

Limitations

This study has several limitations. First, it includes only a small num- ber of participants and one-year follow-up during intervention; future research should validate the findings with larger samples. Second, the ECG was performed after the triage P-score was assigned. Thus, the ECG results were not incorporated into the triage classification and may have hindered accurate diagnosis. Obviously, a more appropriate triage score requires that the ECG examination be done before the triage P-scale is classified; however, this strategy might extend wait times for the triage assessment itself. Third, Patient’s cases were summarized and discussed via an internal mail network. This method was found lacking in providing adequate improvement. A more thorough and methodo- logical approach rather than a mail network may lead to bias minimiza- tion and improved triage to treatment process. Finally, we collected the data retrospectively and examined the cases after the final diagnosis of STEMI. We could not determine which patients deteriorated in the ED. Therefore, we could not follow their process sequence in the ED. Also, we could not identify whether patients had a previous evaluation, for example ECG, before reaching the ED.

Conclusions

Interventions aimed to reduce door-to-balloon time in patients un- dergoing PCI for STEMI can improve patient outcomes. Results indicate a significant improvement in critical quality indicators, namely DTBT, time to triage, and ED assessment factors. The fast-track program for pa- tients with chest pain in ED provides early diagnosis of STEMI and short- enED wait times for coronary reperfusion catheterization. However, the intervention showed expansion of the gaps in three groups of patients including minorities, females and patients with low SES. These findings have significant implications for quality of care for patients with chest pain admitted to the ED. As part of the oath all healthcare providers take it is important to act for bias reduction as part of any gold standard of care protocol. Thus, special attention is needed in order to explore the reason for such delay and in order to improve patient-based approach that may contribute to reducing the time-lags in patients with STEMI.

Author contributions

All authors interpreted the data and edited and approved the final article. MS and TS drafted and conceived the study. MS and TS designed the intervention. MS, TS, RS and AD analyzed the data, designed the

study and performed data collection. MS and TS take responsibility for the paper as a whole.

Role of funding sources

No financial support was received for this research.

Declaration of Competing Interest

There is no conflict of interest.

Appendix 1

Appendix 2.

References

1. Murphy A, Hamilton G, Andrianopoulos N, Yudi MB, Farouque O, Duffy SJ, et al. One- year outcomes of patients with established coronary artery disease presenting with acute coronary syndromes. Am J Cardiol 2019. https://doi.org/10.1016/J.AMJCARD. 2019.01.037.
2. Yeh R, Sidney S, Med MC-NEJ. Preliminary evidence on the effectiveness of preven- tion and the role of population management in acute myocardial infarction. Turner- WhiteCom; 2010 undefined. [n.d.].
3. Liebetrau C, Szardien S, Rixe J, Woelken M, Rolf A, Bauer T, et al. direct admission versus transfer of AMI patients for Primary PCI. Clin Res Cardiol 2011;100:217-25. https://doi.org/10.1007/s00392-010-0231-x.
4. Jernberg T, Johanson P, Held C, Jama BS. Association between adoption of evidence- based treatment and survival for patients with ST-elevation myocardial infarction; 2011 undefined. [JamanetworkCom n.d.].
5. Wright RS, Anderson JL, Adams CD, Bridges CR, Casey DE, Ettinger SM, et al. 2011 ACCF/AHA focused update of the guidelines for the management of patients with un- stable angina/non-ST-elevation myocardial infarction (updating the 2007 guideline). J Am Coll Cardiol 2011;57:1920-59. https://doi.org/10.1016/j.jacc.2011.02.009.
6. Biswas S, Andrianopoulos N, Duffy SJ, Lefkovits J, Brennan A, Walton A, et al. Impact of socioeconomic status on clinical outcomes in patients with ST-segment-elevation myocardial infarction. Circ Cardiovasc Qual Outcomes 2019;12. https://doi.org/10. 1161/CIRCOUTCOMES.118.004979.
7. Bradley EH, Herrin J, Wang Y, Barton BA, Webster TR, Mattera JA, et al. Strategies for reducing the door-to-balloon time in acute myocardial infarction. N Engl J Med 2006;355:2308-20. https://doi.org/10.1056/NEJMsa063117.
8. O’Gara PT, Kushner FG, Ascheim DD, Casey DE, Chung MK, de Lemos JA, et al. 2013 ACCF/AHA guideline for the management of ST-elevation myocardial infarction: ex- ecutive summary. J Am Coll Cardiol 2013;61:485-510. https://doi.org/10.1016/j.jacc. 2012.11.018.
9. Chung S-C, Gedeborg R, Nicholas O, James S, Jeppsson A, Wolfe C, et al. Acute myo- cardial infarction: a comparison of short-term survival in national outcome registries

   in Sweden and the UK. Lancet 2014;383:1305-12. https://doi.org/10.1016/S0140- 6736(13)62070-X.

   Widimsky P, Wijns W, Fajadet J, de Belder M, Knot J, Aaberge L, et al. Reperfusion therapy for ST elevation acute myocardial infarction in Europe: description of the current situation in 30 countries. Eur Heart J 2010;31:943-57. https://doi.org/10. 1093/eurheartj/ehp492.

10. Saban M, Zaretsky L, Patito H, Salama R, Darawsha A. Round-off decision-making: why do triage nurses assign STEMI patients with an average priority? Int Emerg Nurs 2018. https://doi.org/10.1016/J.IENJ.2018.07.001.
11. Takakuwa KM, Burek GA, Estepa AT, Shofer FS. A method for improving arrival- to-electrocardiogram time in emergency department chest pain patients and the effect on door-to-balloon time for ST-segment elevation myocardial infarc- tion. Acad Emerg Med 2009;16:921-7. https://doi.org/10.1111/j.1553-2712. 2009.00493.x.
12. Atzema CL, Schull MJ, Austin PC, Tu JV. temporal changes in emergency department triage of patients with acute myocardial infarction and the effect on outcomes. Am Heart J 2011;162:451-9. https://doi.org/10.1016/J.AHJ.2011.05.015.
13. Sanders SF, DeVon HA. Accuracy in ED triage for symptoms of acute myocardial in- farction. J Emerg Nurs 2016;42:331-7. https://doi.org/10.1016/J.JEN.2015.12.011.
14. Peterson MC, Syndergaard T, Bowler J, Doxey R. A systematic review of factors predicting door to balloon time in ST-segment elevation myocardial infarction treated with percutaneous intervention. Int J Cardiol 2012;157:8-23. https://doi. org/10.1016/J.IJCARD.2011.06.042.
15. Armstrong PW, Gershlick AH, Goldstein P, Wilcox R, Danays T, Lambert Y, et al. Fibri- nolysis or primary PCI in ST-segment elevation myocardial infarction. N Engl J Med 2013;368:1379-87. https://doi.org/10.1056/NEJMoa1301092.
16. Greenland P, Alpert JS, Beller GA, Benjamin EJ, Budoff MJ, Fayad ZA, et al. 2010 ACCF/ AHA guideline for assessment of cardiovascular risk in asymptomatic adults. J Am Coll Cardiol 2010;56:e50-103. https://doi.org/10.1016/j.jacc.2010.09.001.
17. Lin CB, Peterson ED, Smith EE, Saver JL, Liang L, Xian Y, et al. Patterns, predictors, var- iations, and temporal trends in emergency medical service hospital Prenotification for acute ischemic stroke. J Am Heart Assoc 2012;1. https://doi.org/10.1161/JAHA. 112.002345.
18. van der Linden MC, Meester BEAM, van der Linden N. Emergency department crowding affects triage processes. Int Emerg Nurs 2016;29:27-31. https://doi.org/ 10.1016/J.IENJ.2016.02.003.
19. Moskop JC, Sklar DP, Geiderman JM, Schears RM, Bookman KJ. Emergency depart- ment crowding, part 1–concept, causes, and moral consequences. Ann Emerg Med 2009;53:605-11. https://doi.org/10.1016/J.ANNEMERGMED.2008.09.019.
20. Pines JM, Pollack CV, Diercks DB, Chang AM, Shofer FS, Hollander JE. The association between emergency department crowding and adverse cardiovascular outcomes in patients with chest pain. Acad Emerg Med 2009;16:617-25. https://doi.org/10. 1111/j.1553-2712.2009.00456.x.
21. Adriaenssens J, De Gucht V, Van Der Doef M, Maes S. Exploring the burden of emer- gency care: predictors of stress-health outcomes in emergency nurses. J Adv Nurs 2011;67:1317-28. https://doi.org/10.1111/j.1365-2648.2010.05599.x.
22. Forsgren S, Forsman B, Carlstrom ED. Working with Manchester triage – job satisfac- tion in nursing. Int Emerg Nurs 2009;17:226-32. https://doi.org/10.1016/J.IENJ. 2009.03.008.
23. Fallis WM, McMillan DE, Edwards MP. Napping during night shift: practices, prefer- ences, and perceptions of critical care and emergency department nurses. Crit Care Nurse 2011;31:e1-11. https://doi.org/10.4037/ccn2011710.
24. Sorensen NA, Neumann JT, Ojeda F, Schafer S, Magnussen C, Keller T, … Blankenberg

    S. J Am Heart Assoc 2018;7(6):e007297.

    Ting HH, Bradley EH, Wang Y, Nallamothu BK, Gersh BJ, Roger VL, et al. Delay in pre- sentation and reperfusion therapy in ST-elevation myocardial infarction. Am J Med 2008;121:316-23. https://doi.org/10.1016/J.AMJMED.2007.11.017.

25. Miller AL, Simon D, Roe MT, Kontos MC, Diercks D, Amsterdam E, et al. Comparison of delay times from symptom onset to medical contact in blacks versus whites with acute myocardial infarction. Am J Cardiol 2017;119:1127-34. https://doi.org/10. 1016/J.AMJCARD.2016.12.021.
26. Agarwal S, Garg A, Parashar A, Jaber WA, Menon V. Outcomes and resource utiliza- tion in ST-elevation myocardial infarction in the United States: evidence for socio- economic disparities. J Am Heart Assoc 2014;3. https://doi.org/10.1161/JAHA.114. 001057.
27. Haklai Z, Meron J, Applebaum Y, Aburbeh M, Shlichkov G. State of Israel Ministry of Health H e a l t h I n f o r m a t i o n D i v i s i o n emergency room visits; 2014.
28. LEMESHOW S, HOSMER DW. A review of goodness of fit statistics for use in the de- velopment of logistic regression models1. Am J Epidemiol 1982;115:92-106. https://doi.org/10.1093/oxfordjournals.aje.a113284.
29. Halabi S, Elias A, Goldberg M, Hurani H, Darawsha H, Shachar S, et al. Improving door-to-balloon time of patients with ST-Segment Elevation Myocardial Infarction (STEMI) in the emergency department. Isr Med Assoc J 2018;20:476-9.
30. Israel Central Bureau of Statistics. Statistical abstract of Israel 2018 2018. http:// www.cbs.gov.il/reader/shnatonenew_site.htm; 2018, Accessed date: 9 November

    2018.

    Goransson KE, Ehnfors M, Fonteyn ME, Ehrenberg A. Thinking strategies used by registered nurses during emergency department triage. J Adv Nurs 2008;61: 163-72. https://doi.org/10.1111/j.1365-2648.2007.04473.x.

31. Song S, Fonarow GC, Olson DM, Liang L, Schulte PJ, Hernandez AF, et al. Association of Get With The Guidelines-Stroke Program Participation and Clinical Outcomes for Medicare beneficiaries With Ischemic Stroke. Stroke 2016;47:1294-302. https://doi. org/10.1161/STROKEAHA.115.011874.
32. Zahler D, Lee-Rozenfeld K, Ravid D, Rozenbaum Z, Banai S, Keren G, et al. Relation of lowering door-to-balloon time and mortality in ST segment elevation myocardial in- farction patients undergoing percutaneous coronary intervention. Clin Res Cardiol 2019:1-6. https://doi.org/10.1007/s00392-019-01438-6.
33. Yiadom MYAB, Baugh CW, McWade CM, Liu X, Song KJ, Patterson BW, et al. Perfor- mance of emergency department screening criteria for an early ECG to identify ST- segment elevation myocardial infarction. J Am Heart Assoc 2017;6. https://doi.org/ 10.1161/JAHA.116.003528.
34. Mochari-Greenberger H, Xian Y, Hellkamp AS, Schulte PJ, Bhatt DL, Fonarow GC, et al. Racial/ethnic and sex differences in emergency medical services transport among hospitalized US stroke patients: analysis of the National Get With The Guidelines-Stroke registry. J Am Heart Assoc 2015;4. https://doi.org/10.1161/ JAHA.115.002099.
35. Aparicio HJ, Carr BG, Kasner SE, Kallan MJ, Albright KC, Kleindorfer DO, et al. racial disparities in intravenous recombinant tissue plasminogen activator use persist at primary stroke centers. J Am Heart Assoc 2015;4:e001877. https://doi.org/10. 1161/JAHA.115.001877.
36. Pollack CV, Schreiber D, Goldhaber SZ, Slattery D, Fanikos J, O’Neil BJ, et al. Clinical characteristics, management, and outcomes of patients diagnosed with acute pul- monary embolism in the emergency department. J Am Coll Cardiol 2011;57: 700-6. https://doi.org/10.1016/j.jacc.2010.05.071.
37. Schrader CD, Lewis LM. Administration of Emergency Medicine Racial disparity IN EMERGENCY DEPARTMENT TRIAGE; 2013. https://doi.org/10.1016/j.jemermed. 2012.05.010.
38. Blair IV, Steiner JF, Havranek EP. Unconscious (implicit) bias and health disparities: where do we go from here? Perm J 2011;15:71-8.
39. Fitzgerald C, Hurst S. Implicit bias in healthcare professionals: a systematic review. BMC Med Ethics 2017;18. https://doi.org/10.1186/s12910-017-0179-8.
40. Burgess D, Van Ryn M, Dovidio J, Saha S. Reducing Racial bias among health care pro- viders: lessons from social-cognitive psychology. J Gen Intern Med 2007;22:882-7. https://doi.org/10.1007/s11606-007-0160-1.
41. Kronick SL, Kurz MC, Lin S, Edelson DP, Berg RA, Billi JE, et al. Part 4: Systems of Care and Continuous Quality Improvement. Circulation 2015;132:S397-413. https://doi. org/10.1161/cir.0000000000000258.
42. Elf M, Fro P, Lindahl G, Wijk H. shared decision making in designing new healthcare environments-time to begin improving quality. BMC Health Serv Res 2015;15:1-7. https://doi.org/10.1186/s12913-015-0782-7.
43. Bradley EH, Herrin J, Wang Y, McNamara RL, Webster TR, Magid DJ, et al. racial and ethnic differences in time to acute reperfusion therapy for patients hospitalized with myocardial infarction. JAMA 2004;292:1563. https://doi.org/10.1001/jama.292.13. 1563.
44. Weintraub NL, Collins SP, Pang PS, Levy PD, Anderson AS, Arslanian-Engoren C, et al. Acute heart failure syndromes: emergency department presentation, treatment, and disposition: current approaches and future aims. Circulation 2010;122:1975-96. https://doi.org/10.1161/CIR.0b013e3181f9a223.
45. Arslanian-Engoren C, Patel A, Fang J, Armstrong D, Kline-Rogers E, Duvernoy CS, et al. Symptoms of men and women presenting with acute coronary syndromes. Am J Cardiol 2006;98:1177-81. https://doi.org/10.1016/J.AMJCARD.2006.05.049.
46. Fournier S, Muller O, Ludman AJ, Lauriers N, Eeckhout E. Influence of socioeconomic factors on delays, management and outcome amongst patients with acute myocar- dial infarction undergoing primary percutaneous coronary intervention. Swiss Med Wkly 2013;143:13817. https://doi.org/10.4414/smw.2013.13817.
47. Goldman S, Radomislensky I, Ziv A. KP-I journal of The impact of neighborhood so- cioeconomic disparities on injury. Springer; 2018 undefined. [n.d.].
48. Cenko E, Yoon J, Kedev S, … GS-J internal. Sex differences in outcomes after STEMI:

    Effect modification by treatment strategy and age. JamanetworkCom; 2018 unde-

    fined. [n.d.].