Hospital Rates of Thrombolysis for Acute Ischemic Stroke
The Influence of Organizational Culture
Background and Purpose— The purpose of this study was to determine if organizational culture explains differences in rates of intravenous thrombolysis for acute ischemic stroke between different hospitals.
Methods— A cohort study was done in 12 centers admitting 5515 consecutive patients with acute stroke in The Netherlands. A multilevel logistic regression model was used to relate the likelihood of treatment with thrombolysis to characteristics of the organizational culture of the centers. Organizational culture was defined by 10 characteristics and scored by a panel. A sum score was created by adding all scores and dividing by 10.
Results— Thrombolysis rates varied from 5.7% to 21.7%. We observed an association between thrombolysis and the availability of informal and formal feedback (OR, 1.18; 95% CI, 1.09 to 1.28); a learning culture (OR, 1.12; 95% CI, 1.02 to 1.23); uncompromising, individual clinical leadership (OR, 1.12; 95% CI, 1.03 to 1.23); explicit goals (OR, 1.08; 95% CI, 1.01 to 1.17); and with the sum score (OR, 1.12; 95% CI, 1.02 to 1.23).
Conclusions— Several cultural characteristics of the hospital organization are related to thrombolysis rate. Organizational culture may be an important target for interventions aimed at increasing rates of thrombolysis for acute ischemic stroke in hospitals.
In most hospitals, only 2% to 10% of all admitted patients with stroke are treated with thrombolysis, although 25% might be eligible for treatment.1 Research on improving delivery of thrombolysis has been focused primarily on characteristics of patients with stroke. Some attention has been paid to structural characteristics of the organization such as the availability of protocols and training,2 but none at all to the influence of organizational culture. In this study, we assessed the association between thrombolysis rates in hospitals and organizational cultural characteristics.
This study was designed as a cohort study in 12 centers covering 11% of all hospitals in The Netherlands. At the start of the study, the participating hospitals had a mean thrombolysis rate of 5% (range, 0% to 10%), similar to the mean thrombolysis rate in The Netherlands at that time.
We used a mixed methods approach with both quantitative and qualitative research methods based on a Delphi approach.3 During the study period of 2 years, all patients >18 years of age with acute stroke were included and clinical characteristics that might be related to the delivery of thrombolysis were recorded. The primary outcome in all registered patients was treatment with thrombolysis or not. On hospital level, we calculated a thrombolysis rate by dividing the number of patients with stroke treated with thrombolysis by the total number of patients with stroke admitted during the inclusion period.
We identified 8 cultural characteristics from a qualitative case study on improving door-to-balloon times for patients with acute myocardial infarction.4 We added 2 additional cultural characteristics that might be related to (lack of) resistance to change: unanimous partnership and cooperative partnership5,6 (Supplemental Table I; available at http://stroke.ahajournals.org).
During the study period, a vascular neurologist in each center kept a diary of their activities to improve the rate of thrombolysis and onset-to-needle time. These diaries were used as input for “face-to-face” interviews with neurologists and telephone interviews with a stroke nurse in each center conducted by a neurologist and an organizational scientist (Supplemental Table II⇓).
The data were used to attribute a score between zero and 10 for each hospital on each of the 10 characteristics. A sum score was made for each center by adding up all scores for each characteristic and dividing by 10.
The full transcripts of both interviews were made anonymous and were analyzed independently by 2 raters (R.H. and I.N.F.) with a modified Delphi approach. If after 3 rounds no consensus could be reached, a third rater (J.D.H.v.W.) was involved to tip the balance. As an example, a full case description of 2 hospitals is available in Supplemental Table III.
For the analysis of the association between treatment with thrombolysis and several cultural characteristics, we used a multilevel logistic regression model to be able to adjust for the potential clustering effect. If the analysis showed a relevant association, we subsequently adjusted for hospital size and teaching facilities.
Of the 5515 registered patients with stroke, 701 (12.7%) were treated with intravenous recombinant tissue plasminogen activator. Thrombolysis rates varied from 5.7% to 21% (Table 1). There were no significant associations between patient characteristics and cultural characteristics.
One center acted as an outlier and was omitted from further analyses. The unadjusted multilevel logistic analysis showed a significant association between thrombolysis rate and several cultural characteristics. A statistically significant association between “informal and formal feedback,” “learning culture,” “uncompromising, individual clinical leadership,” and “explicit goals” and the likelihood of receiving thrombolysis was observed (Table 2). Also, the overall sum score of cultural characteristics was associated with thrombolysis. Adjustments for hospital size and teaching versus nonteaching hospital did not change the size and direction of these associations. When these characteristics (minus the sum score) were combined into one multivariable multilevel logistic regression model, only “feedback” showed a significant association with thrombolysis rate (OR, 1.19; CI, 1.04 to 1.36). Increases in thrombolysis rate were not associated with an increase in nonadherence to protocols or occurrence of adverse events.
This study shows that the availability of certain cultural characteristics increases the likelihood of receiving thrombolysis for patients. Based on our calculations, a reasonable improvement in cultural sum score of ½ SD might lead absolute to increases of 1% in thrombolysis rates. This makes organizational culture an interesting target for interventions aimed at improving thrombolysis rates.
Our results suggest that centers need to have explicit, shared goals concerning door-to-needle time and thrombolysis rate. These should be monitored continuously and feedback should be regularly provided. Clinical leaders need to be identified, appointed, and/or trained who are respected by their peers, who both inspire and push “individuals and the organization to achieve a high standard of care.”7
We extend our gratitude to all stroke care teams who participated in this study and to the PRACTISE investigators. The PRACTISE investigators are listed in the study protocol: Dirks M, Niessen LW, Huijsman R, van Wijngaarden J, Minkman MM, Franke CL, van Oostenbrugge RJ, Koudstaal PJ, Dippel DW; Promoting Acute Thrombolysis for Ischaemic Stroke (PRACTISE). Int J Stroke. 2007;2:151–159.
Source of Funding
This study was funded by ZonMW, the Dutch center for health care research (Grant No 945-14-217).
- Received June 3, 2009.
- Revision received July 7, 2009.
- Accepted July 9, 2009.
Kwan J, Hand P, Sandercock P. Improving the efficiency of delivery of thrombolysis for acute stroke: a systematic review. QJM. 2004; 97: 273–279.
Jones R. Why do qualitative research? BMJ. 1995; 311: 2.
Bradley EH, Curry LA, Webster TR, Mattera JA, Roumanis SA, Radford MJ, McNamara RL, Barton BA, Berg DN, Krumholz HM. Achieving rapid door-to-balloon times: how top hospitals improve complex clinical systems. Circulation. 2006; 113: 1079–1085.
Schouten LMT, Hulscher MEJL, Akkermans R, van Everdingen JJE, Grol RPTM, Huijsman R. Factors that influence the stroke care team’s effectiveness in reducing the length of hospital stay. Stroke. 2008: 2525–2521.
Carroll JS, Edmondson AC. Leading organisational learning in health care. Qual Saf Health Care. 2002; 11: 51–56.