Teamwork and Safety Culture in Small Rural Hospitals in Mississippi
November / December 2006
Feature Article
Teamwork and Safety Culture in Small Rural Hospitals in Mississippi
This study examines attitudes towards safety and teamwork in eight rural hospitals in Mississippi. While studies have focused on attitudes of hospital workers towards patient safety and teamwork in urban areas and/or specialized units of care (USDA, 2003; Ricketts et al., 1999; IOM, 2005), few studies measure such attitudes in rural areas. Given the paucity of research on rural healthcare, findings from our study will help define the role of safety and teamwork in improving patient care in rural hospitals.
Even though more than 59 million Americans live in rural areas, little is known about patient safety issues and challenges facing rural healthcare providers (Robinson & Guidry, 2001). In general, prior research has focused on challenges in rural healthcare such as demographic factors, staff recruitment barriers, lower reimbursement rates, greater percentage of uninsured patients, lack of subspecialty care, and geographic isolation (Larson & Hill, 2005; Wright, 2001; Coombs, 2001; Baldwin, et al., (2004). In addition, there is a perception that rural hospitals provide substandard care. To a certain degree, prior research has supported the belief that rural hospitals lag behind their urban counterparts in evidence-based best practices, technology, and qualified personnel. Findings on safety practices within rural hospitals, however, have been inconclusive. For example, Rieber et al. (1996) found that patients in rural areas were at higher risk for certain types of adverse events, and Radcliff et al. (2003) found that rural hospitals had lower rates for various indicators of unsafe practices. Moreover, research on rural healthcare has been limited by sample size, data availability, and lack of quality assessment measures. This study uses multiple hospital sites, surveys all direct care providers within these hospitals, and uses a standardized, validated quality measurement tool to assess safety and teamwork attitudes.
Methods
Data Collection
Data for this paper were collected from eight rural hospitals in Mississippi. Each hospital is at least 45 miles from a Standard Metropolitan Statistical Area. The number of beds per hospital ranged from 9 to 82. Six of the eight hospitals have a critical care access designation, having fewer than 25 beds. Five of the hospitals are located in the Mississippi Delta. More than 60% of the population is African-American, and the per capita income is less than $12,000 per year. Greater than 30% of the region’s population lives below the poverty level. For comparative purposes, the remaining three hospitals are located in east central Mississippi. This area has a larger, albeit equally impoverished, Caucasian population. This area’s per capita income is approximately $14,000, with 20% of the population below the poverty level (Mississippi Development Authority, 2004-2005; 2003).
All healthcare providers with direct patient contact were eligible to participate in the study. Surveys were distributed prior to a continuing education class on patient safety attended by 192 of the 219 providers at the eight hospitals (85%). Of those who received a survey, 98.9% (N=190) completed and returned it. The number of surveys received from each hospital ranged from 12 to 66.
Survey Instrument
The short version of the Teamwork and Safety Climate Safety Attitude Questionnaire, (SAQ) which contains 34 items rather than the full 62 of the original version, was used to collect data (Sexton et al., in press). Given the time and personnel limitations often faced by rural healthcare providers, and, in order to achieve a high response rate, this instrument was selected because of its brevity and ease of completion. The instrument was initially developed for use in intensive care units within a large urban academic center and has been successfully administered in outpatient settings, general inpatient wards, and operating rooms.
Demographic data were collected, including job title, gender, age, ethnicity, shift, and number of years in specialty and in organization. Job title is separated into three categories: physician (n=23), nurse (n=134), and other (n=26). The “other” category includes pharmacist, unit clerk, and respiratory therapist. Initially, pharmacist was to be a separate category; however, only four pharmacists are employed by the eight hospitals. Age is separated into three categories: younger than 35 (n=43), 35 to 44 (n=58), and 45 and older (n=83). Ethnicity is separated into two categories: white (n=127) and non-white (n=44). The “non-white” category includes African-American (n=40), Hispanic (n=1), and Asian (n=3). Shift is separated into two categories: day (n=111) and evening/night (n=71). The number of years in specialty and at organization were separated into four categories: less than 2 years (n=29; n=49), 2 to 7 years (n=40; n=49), 8 to 20 years (n=68; n=48), and over 20 years (n=40; n=24).
Results
Factor Analysis
All 34 variables from the short version of the Teamwork and Safety Climate Safety Attitude Questionnaire (SAQ) were included in the exploratory factor analysis. The initial factor analysis yielded eight significant factors (Eigen values that exceeded 1.0). However, only three factors had alpha coefficients that exceeded 0.70: teamwork collaboration, patient safety culture, and teamwork knowledge transfer. The factor structure previously employed by Sexton et al. (in press), was acceptable for urban areas, but we believed that the new structure more accurately reflected concepts important to rural hospital healthcare providers. Only variables that had a factor loading that exceeded 0.45 were included on a particular scale. The teamwork knowledge transfer factor combines variables from the patient safety climate and the teamwork scales. Identifying the teamwork knowledge transfer factor is important because it delineates the close relationship between teamwork and creating a patient safety culture of open communication.
Teamwork and Patient Safety Climate
Means, standard deviations, and standard errors of the means for each of the variables within the three factors with alpha coefficients that exceed 0.70 are reported in Table 1. Findings generally showed positive attitudes toward the value of teamwork, management’s role in creating a culture of patient safety, and the importance of teamwork in the transfer of knowledge and communication between healthcare professionals. The mean score ranged from 4.31 (“It is easy for personnel here to ask questions when there is something they do not understand”) to 3.67 (“The culture in the clinical area makes it easy to learn from the errors of others”) on a 5-point Likert scale.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Provider Differences
ANOVA was conducted to assess possible differences between the three factors and the following provider group characteristics: provider position, shift, gender, ethnicity, age, number of years of working in the specialty area, and number of years of working in the organization. The results showed that no significant differences existed with any of the three factors by gender, ethnicity, age, number of years working in the specialty area, or number of years working in the organization. Significant differences did exist by shift (day vs. evening/night) on the patient safety culture factor (1, 168; F=4.66; p=0.032) and on the teamwork collaboration factor (1, 167; F=5.95; p=0.016). Similarly, a significant difference existed by position on the teamwork collaboration factor (2, 167; F=5.21; p=0.006). Using the Bonferroni test to examine nested differences between employment positions, we found that the “other” category, e.g. pharmacists, ward clerks, and respiratory therapists, is not significantly different from either the physician or nurse groups with respect to teamwork knowledge transfer. But, within this same factor, there is a significant difference between physicians and nurses (p=0.008) in which physicians were more likely to have positive attitudes about teamwork, while nurses were less likely to have positive teamwork attitudes.
Statistically significant differences existed only between shift and provider position on the teamwork collaboration and the teamwork knowledge transfer scales. Figures 1 and 2 provide a graphic representation of the magnitude of these differences. Those working during the day shift had significantly more positive beliefs in management’s role in helping make the organization safe and improving the safety culture. The day shift also felt more positive regarding collaborative efforts and open communication between healthcare providers. Physicians had a more positive view than nurses of collaborative efforts within the organization. Despite these differences, the evening/night shift personnel and nurses did have positive perceptions concerning management’s role in creating and maintaining a patient safety culture and on the value of teamwork, collaboration, and open communication between healthcare providers.
Discussion
Findings from our analysis provide new insights into healthcare providers’ perceptions of safety culture and the importance of teamwork at small rural hospitals. This is the first study to examine rural healthcare provider perceptions of team work and safety at a system level. As a result of the rural locations from which the data were obtained, our findings differ from prior findings from urban and suburban areas in three ways.
First, healthcare providers at small rural hospitals have more positive attitudes toward teamwork and management’s role in creating a safe environment. In general, the mean scores for each question were higher than have been reported in previous studies using the SAQ (Sexton et al., in press; Weingart et al., 2004; Sorra & Nieva, 2004). Healthcare providers were satisfied with the quality of collaboration, felt suggestions concerning safety were acted upon by management, and felt that leadership was driving their hospitals to be safety centered.
Second, findings from this study show that minimal differences existed across economic and demographic characteristics between healthcare providers on all three factors: teamwork collaboration, patient safety culture, and teamwork knowledge transfer. Where statistically significant differences did exist, responses ranged from “positive” to “very positive.” Although physicians had a more positive attitude toward existing collaborative efforts, the mean score for nurses on the teamwork collaborative scales was 3.96 out of 5 as compared to 4.48 for physicians.
Third, the factor structure from the rural hospitals is different from the factor structure from previous studies. Our research identified three significant factors. Unlike prior research using the SAQ in which the safety and teamwork questions loaded on separate factors, in our study, questions from both the teamwork and safety sections significantly loaded on a single factor. That is, healthcare providers at rural hospitals were able to link the concept of teamwork to the transfer of knowledge, thereby creating a patient safety culture of open communication. For example, the statement “important issues are well communicated at shift changes” loaded on the same factor as the statement “I am satisfied with the quality of collaboration that I experience with nurses in this clinical area.”
Data Limitations
To date, little research on teamwork and patient safety in rural areas has been conducted. Even though our findings provide new insights into small rural hospital culture, there are three primary limitations. First, data for this study were collected in one state, Mississippi. Given the economic and demographic characteristics of Mississippi, extrapolating these findings to other rural settings and different geographic locations must be done with caution. Second, we are unable to assess individual hospital effects or differences on the short version of the Teamwork and Safety Climate Safety Attitude Questionnaire (SAQ) due to the small “n” at each hospital. Consequently, differences at a hospital level may exist. The third limitation is related to the short version of the SAQ instrument itself. Having originated in the aviation industry and being further developed primarily for use in ICUs, the short version of the SAQ has not been validated as a patient safety and teamwork survey instrument for rural healthcare.
Conclusion
Findings from this research provide a starting point for understanding patient safety culture issues in rural areas. These data seem to suggest that healthcare providers in rural areas have positive attitudes toward teamwork and safety and are able to link these two attitudes into their daily work environment to create a culture that endorses patient safety. Whether these findings are simply an artifact of our research or whether these findings can be extrapolated to other areas requires further investigation.
Bill Rudman is a professor in health information management and medicine and the director of the Clinical Health Sciences Graduate Program at the University of Mississippi Medical Center. He may be contacted at brudman@shrp.umsmed.edu. Andrew Brown is a professor at the University of Mississippi Medical School. Rudman and Brown have worked in the area of patient safety, specifically on capturing medication errors in ambulatory rural settings. Most recently, their work is developing a regional health information organization in rural Mississippi. Also at the University of Mississippi Medical Center, Jessica Bailey is assistant professor and Ann Peden is associate professor in the Department of Health Information Management. Paula Garrett is associate professor and writing program director At Millsaps College in Jackson, Mississippi. Eric Thomas is associate professor of medicine at the University of Texas Medical School at Houston.
References
Baldwin L., MacLehose, R. F., Hart, L. G., Beaver, S. K., Every, N., & Chan, L. (2004). Quality of care for acute myocardial infarction in rural and urban U.S. hospitals. Journal of Rural Health, 20(2), 99-108.
Coombs, J. B. (2001). Quality of care in rural settings: Bringing the “new quality” to rural practice. In J.P. Geyman, T.E. Norris, and L.G. Hart (Eds.), Textbook of rural medicine (pp. 325-340). New York: McGraw-Hill.
Institute of Medicine (IOM) Committee on the Future of Rural Health Care, Board on Health Care Services. (2005). Quality through collaboration: The future of rural health. Washington, DC: National Academies Press.
Larson, S. L., & Hill, S. C. (2005). Rural-urban differences in employment-related health insurance. Journal of Rural Health, 21(1), 21-30.
Mississippi Development Authority. Resources. Available at http://www. mda.state.ms.us/content.aspx?url=/page/resources&
Ricketts, T. C. III, Johnson-Webb, K. D., & Randolph, R. K. (1999). Populations and places in rural America. In T.C. Ricketts, III (Ed.), Rural health in the United States (pp. 7-24). New York: Oxford University Press.
Robinson, J. III, & Guidry, J. J. (2001). Recruiting, training, and retaining rural health professionals. In S. Loue and B.E. Quill (Eds.), Handbook of rural health (pp. 337-353). New York: Kluwer Academic/Plenum Publishers.
Sexton, J. B., Helmreich, R. L., Neilands, T. B., Rowan, K., Vella, K., Boyden, J., Roberts, P. R., & Thomas, E. J. (In press). The safety attitudes questionnaire: Psychometric properties, benchmarking data, and emerging research. BMC Health Services Research.
Sorra, J. S., & Nieva, V. F. (2004, September). Hospital survey on patient safety culture. (Prepared by Westat, under Contract No. 290-96-0004). AHRQ Publication No. 04-0041. Rockville, MD: Agency for Healthcare Research and Quality.
U. S. Department of Agriculture (USDA). Economic Research Service. (2003). Measuring rurality: What is rural? www.ers.usda.gov/briefing/rurality/whatisrural.
Weingart, S. N., Farbstein, K., Davis, R. B., & Phillips, R. S. (2004). Using a multihospital survey to examine the safety culture. Joint Commission Journal of Quality and Safety, 30, 125-132.
Wright, G. E. (2001). The economics of rural practice. In J.P. Geyman, T.E. Norris, and L.G. Hart (Eds.), Textbook of Rural Medicine, (pp. 275-288). New York: McGraw-Hill.