|Home | About | Journals | Submit | Contact Us | Français|
A healthy safety culture is integral to positive health care. A sound safety climate is required in Obstetrics and Gynecology to prevent adverse outcomes.
The objective of this study was to assess and compare patient safety culture in two departments of Obstetrics and Gynecology.
Using a closed-ended standard version of Hospital Survey on Patient Safety Culture (HSOPS), respondents were asked to answer 42 survey items, grouped into 10 dimensions and two outcome variables in two tertiary care teaching hospitals in Delhi. Qualitative data were compared using Fisher's exact test and chi-square test wherever applicable. Mean values were calculated and compared using unpaired t-test.
The overall survey response rate was 55%. A positive response rate of 57% was seen in the overall perception of patient safety that ranged from very good to acceptable. Sixty-four percent showed positive teamwork across hospital departments and units, while 36% gave an affirmative opinion with respect to interdepartmental handoffs. However, few adverse events (0-10) were reported in the last 12 months and only 38% of mistakes by doctors were reported. Half of the respondents agreed that their mistakes were held against them. There was no statistical difference in the safety culture between the two hospitals.
Although the perception of patient safety and standards of patient safety were high in both the hospitals' departments, there is plenty of scope for improvement with respect to event reporting, positive feedback, and nonpunitive error.
Safety culture is a complex phenomenon that consists of subcultures such as leadership, teamwork, evidence-based practices, communication, learning, and patient-centered practices.(1) Initially, the concept of safety culture was practiced in high-risk areas such as aviation, nuclear energy, and shipping, but health care is an equally challenging, dynamic, and potentially high-risk area. Data show that 50% of adverse events in health care are preventable.(2)
The Institute of Medicine stated that health care organizations should develop and promote a safety culture where adverse events are reported without people being blamed, provide scope for improvement to doctors by enabling them to learn from their mistakes, and prevent further errors.(3) To assess patient safety culture, nine surveys are well identified, out of which psychometric testing quantity and quality is comprehensive only in four surveys.(4) One such validated and reliable survey is the Hospital Survey on Patient Safety Culture (HSOPS) developed by the Agency for Healthcare Research and Quality (AHRQ), which was used in our study.(5)
Obstetrics and Gynecology involves a dual high risk of both maternal and fetal morbidity and mortality, and requires a sound safety climate to prevent adverse outcomes. Hence we decided to undertake a pilot survey using HSOPS, to assess and compare the safety culture in the departments of Obstetrics and Gynecology of two teaching tertiary care public hospitals. To the best of our knowledge, this is the first of its kind in a developing country.
The questionnaire survey (HSOPS) was distributed in the departments of Obstetrics and Gynecology of two tertiary care teaching hospitals in Delhi. The hospitals are located in two different zones of the city and are under different administrations. Both hospitals have an annual delivery rate of more than 10,000 and cater to the middle and low socioeconomic strata of patients.
The questionnaire was distributed by the authors to all consultants, senior residents, and postgraduates of the department working at that point of time in the hospitals after their consent. Junior residents, interns, medical students, nurses, and other paramedical staff were not eligible to participate in the survey. Completed forms were collected after 1 week. Those unable to give a reply were given two reminders at 48 h and again after 3 days, after which they were declared nonresponders and excluded from the survey. Names of doctors were not written on the survey form and confidentiality was ensured.
The completeness of the survey forms was assessed. The form was excluded from analysis if it was less than 50% complete or if consecutive answers were similar under two main headings. If there was more than one answer to a question, it was taken as a missing value. The latter was replaced by the mean of responses for that question.
HSOPS is a validated questionnaire that uses 5-point Likert scales of agreement (“Strongly disagree” to “Strongly agree”) or frequency (“Never” to “Always”). It has 10 safety culture dimensions and four outcome variables, for a total of 42 items [Table 1]. The outcome variables are grouped as frequency of event reporting, overall perception of patient safety, patient safety grade, and total number of events reported. The safety culture dimensions at the unit level are expectations and actions of supervisor/manager promoting safety, organizational learning-continuous improvement, teamwork within hospital units, openness in communication, feedback and communication about error, nonpunitive response to error, staffing and hospital management support for patient safety. At the hospital level, the dimensions assessed are teamwork across hospital units and handoffs and transitions. Each dimension has an acceptable reliability (Cronbach's alpha greater than or equal to.60), with reliability coefficients ranging from.63 to.84.(5)
The positive frequency of each response for the survey item was calculated and the missing responses were removed from the denominator. For ease of calculation, out of all 5 responses the lowest 2 response categories (strongly disagree/disagree; never/rarely) and the highest two response categories (strongly agree/agree and most of the time/always) were combined. After the individual calculation of each response, the composite positive frequency of responses on safety culture dimensions were calculated. Qualitative data were compared using Fisher's exact test and chi-square test wherever applicable. Mean values were calculated and compared using unpaired t-test. P< 0.05 was taken as significant.
Out of a total of 170 eligible doctors, 98 completed the survey form. The overall response rate was 55%, and the difference in the two hospitals was not statistically significant (49% vs 60%). Out of 6 excluded forms, 4 were <50% complete, while 2 had the same answer repeated in 20 consecutive items. Hence a total of 93 survey forms were analyzed.
The background variables of all the participants are summarized in Table 2. All the participants confirmed having direct interaction with patients. On comparing the participant information of the two hospitals, there was a significant difference with respect to staff position, work experience, and duty hours.
A positive response rate of 57% was seen in the overall perception of patient safety. The overall rating of patient safety ranged from very good to acceptable. However, only 38% of mistakes by doctors were reported irrespective of their potential to harm the patient. A few adverse events were reported in the last 12 months, ranging 0-10 in both the hospitals. Sixty percent of the participants also noted that the majority of these written reports pertained only to mortality data, while adverse events regarding patient morbidity were seldom reported.
A composite positive response rate of 55% was obtained on analyzing the hospital-wide safety dimension. Out of this, 64% showed positive teamwork across hospital departments and units, while only 36% gave an affirmative opinion with respect to hospital and interdepartmental handoff and transition.
On analyzing at the unit level, an overall positive response rate was 63%. The eight safety culture dimensions are separately tabulated in Table 3. However, certain individual response rates were notable. Fifty-four percent agreed, while 16% of the participants were neutral in their opinion that their mistakes were held against them. On analyzing the response to question “When an event is reported it feels that the person is written up and not the problem,” 53% agreed and 18% were neutral in their opinion. Hence the composite frequency rate of nonpunitive response to error was as low as 39%. With respect to “staffing,” 47% agreed, while 20% were neutral regarding the “crisis mode” action of trying to do too much too quickly. Seveny-eight percent also agreed that they spent longer hours in the hospital than was ideal for patient care.
On comparison of the positive frequency of each response to the HSOPS questionnaire between the two hospitals, no significant statistical difference was observed. The various responses in the sections of questionnaire are summarized in Table 4.
There is a widespread interest in improving patient safety in health care including Obstetrics and Gynecology. Poor communication and teamwork have been identified in almost 50% of maternal deaths and 43% of malpractice claims in obstetrics.(6,7) Getting the “right patient safety culture” is an important component in improving patient safety, which can be assessed by various surveys. We used the HSOPS questionnaire. Studies have shown that it has similar reliability and predictive validity as the safety attitude questionnaire (SAQ). HSOPS safety culture dimensions were the best predictors of frequency of event reporting and overall perception of patient safety, while SAQ and HSOPS dimensions both predicted patient safety grade.(8)
One of the strategies suggested by the Institute of Medicine, USA to improve patient safety was identification and mandatory reporting of incidents.(3) The event reporting, according to the present survey, dealt mainly with maternal mortality, and events of serious morbidity were seldom reported. Various ongoing projects are collecting data on near-miss maternal mortality in our hospital. Newer projects can be undertaken to audit near-miss morbidity on the lines of the UK Obstetric Surveillance System (UKOSS) in order to understand the deficiencies in our own facility.(9)
Almost half of the doctors in our study had an affirmative response to the fact that their mistakes were held against them and that they were held responsible for adverse outcomes. In one of the key components to promote patient safety, the American Congress of Obstetricians and Gynecologists (ACOG) promotes the concept of a “just culture,” which accepts that highly competent doctors can also make mistakes.(10) There should be a correct balance between individual accountability and punishment for an unintended human error.(11) Instead of a punitive response, systems must assure that all staff who report the adverse events are supported and acknowledged for their contribution and are continually encouraged by the knowledge that their reporting has led to safer conditions.(12)
Although teamwork was rated high in our survey (64%), there was still a lacuna in interdepartmental handoffs, in which the positive response rate was only 36%. This is similar to another study on surgeon information transfer and communication, where out of 328 case descriptions, 87 reports and 67 reports were of blurred responsibility and inhibited communication respectively, leading to 31% adverse patient consequences.(13) Obstetricians should develop face-to-face standardized handoff protocols and use structured communication techniques such as Situation, Background, Assessment Recommendation (SBAR) wherever possible.(14)
The two hospital departments involved in the survey cater to different geographical areas and have different administrations. However, it is demonstrated from the survey that safety culture is localized by a specific clinical area rather than by a specific hospital.(15,16) There were certain limitations in the survey. The response rate of the survey was 51%, which, although low, was adequate for evaluation. This could be due to the administration of a complex survey to an extremely busy group of doctors, i.e., obstetricians and gynecologists. In addition, there was a difference in the background variables with respect to staff position, work experience, professional tenure, and working hours. However, the overall results were not affected by this bias as this was an observational study. Another small bias was the time frame in which the questionnaire was returned. Some participants returned it in 1 h while some took 1 week. Such time delays can result in changes in the response.
To conclude, although the perception of patient safety and the standards of patient safety were high in both the hospital departments, there is plenty of scope for improvement with respect to event reporting, positive feedback, and nonpunitive error. Event reporting needs to be improved and standardized, as only a few events were reported according to the survey. The adverse event reporting protocol needs to be improved in order to improve patient management, rather than focusing on individual mistakes. Departmental guidelines should be established and updated periodically in response to adverse events and prospective improvement. There is a need for formal training and simulation programs in techniques not only to improve obstetric skills but also for teamwork behavior, communication, and staff attitudes. There is a need for development and testing of better communication tools to improve handoffs both inter- and intradepartmental. Periodic multi-institutional surveys need to be conducted to create a safe culture.
There are no conflicts of interest.