The effects of patients initiated aggression on Chinese medical students’ career planning

The study was a cross-sectional survey conducted in 2015 in the West China Clinical Medicine School. Overall, there were 180 eligible participants. After being informed about the purpose of the study and providing verbal consent, 180 randomly recruited medical students completed a self-administered questionnaire along with the Short Form-36 (SF-36) scale testing quality of life. 23 (12.8%) refused to participate, and finally 157 were qualified for this study, with an overall response rate of 87.2%.

The questionnaire contained five sections covering demographic characteristically, characteristics and causes of violence experienced during the previous 12 months, and the effects of such violence on the medical students including their quality of life. Demographic characteristics included age, gender, marital status, educational level, work experience in health sectors, mean daily hours at work, weekly night shift, weekly clinic service visits, and monthly discharges. Characteristics of violence during the previous 12 months included extreme violence (physical violence led to surgical treatment or hospitalization), physical violence (students suffered physical damage), mental abuse (patients used language to insult students), offensive threats (patients verbally expressed the intention for physical violence), verbal sexual harassment (patients verbally flirted with students), and physical sexual harassment (patients physically flirted with students). The three dimensions of patient initiated aggression were emotional attack (EA, mental abuse and offensive threat), sexual harassment (SH, verbal and physical sexual harassment), and physical violence (PV, extreme violence and personal attack). Items investigating the occupational plan were added to the patient initiated aggression scale. Causes of violence included insufficient communication (i.e., the victim was a poor communicator; medical students are usually less experienced than physicians, leading to insufficient doctor–patient communication, so the patient may not make the most informed decision about their care, which is likely to cause conflicts), lack of trust between patients and physicians, low-level personal quality (meaning a person who is rude and prone to address problems by violence and does not respect people; put another way, this is a problem of the perpetrator lacking a humanistic quality), staff attitudes, unmet expectations of patients/families, waiting time, out-of-pocket fees, patient’s death, emergency circumstances, lack of medicines or needed services, and others. These were fixed choices. We read some papers containing questionnaire designs relating to our study (Health-care personnel and patient initiated aggression in the emergency departments of a volatile metropolis: results from Karachi, Pakistan [8]; Workplace towards workers in the emergency departments of Palestinian hospitals: a cross-sectional study [9]; and Chinese doctors are under threat [12], which we have cited in our paper). Effects of violence on medical students included pursuing a medical career, choosing a non-medical career, and giving up a medical career. Medical students’ quality of life was measured using the SF-36 scale, comprehensively summarizing respondents’ survival quality based on eight aspects: physical functioning, role physical, bodily pain, general health perceptions, vitality, social role functioning, role emotional, and mental health. The SF-36 scale had high consistency with a Cronbach’s α coefficient greater than 0.7 (except for social role functioning and vitality).

The survey response rate for the questionnaire and SF-36 form was 87.2%. Participants predominantly were 20–24 years old (66.3%), female (56.1%), and single (96.2%) and worked 6–8 h daily (54.8%) with almost all having night shifts (96.2%) (Table 1).

Of the 157 medical students surveyed, 48 (30.6%) reported having suffered patient initiated aggression at least once during the previous year in the form of mental abuse (32, 20.4%), offensive threats (23, 14.6%), PV (13, 8.3%), SH (verbal: 13, 8.3% or physical:3, 1.6%), and extreme violence (PV leading to surgical treatment or hospitalization) (1, 0.6%). Among reasons for violence, insufficient communication was the most common (27.2%) followed by lack of trust between patients and physicians (25.7%), patient’s low-level personal quality (13.2%), staff member’s attitudes (10.6%), unmet expectations of patients/families (6.9%), waiting time (5.6%), out-of-pocket fees (4.3%), and patient’s death (3.5%).

The three dimensions of patient initiated aggression were distributed as follows: EA (41 times, 26.1%), SH (15 times, 9.6%), and PV (13 times, 8.3%). In analyses by gender, age, marital status, educational level, and daily working hours, EA occurrence differed among age groups (χ2 = 9.786, P = 0.020), with participants aged >30 years more likely to suffer EA than participants aged 20–24 years (χ2 = 7.567, P = 0.006) and participants aged 25–29 years (χ2 = 10.106, P = 0.001). Women were more likely exposed to PV (χ2 = 6.796, P = 0.009) (Table 2). Patient initiated aggression appeared not to impact medical students’ career planning over the following three years; most medical students (89.2%) were inclined to continue their medical careers. Only 17 students decided to choose non-medical careers (12.5%) or to give up their medical career (2.1%) (Table 3).

The SF-36 scale scores among medical students by frequency of patient initiated aggression showed that most medical students’ patient initiated aggression score were concentrated in the zero frequency group (69.4%), while there were 48 medical students in the intermediate frequency group (30.6%). Comparing the SF-36 scores in terms of the difference in patient initiated aggression frequency, the score of the low–intermediate frequency group was lower than in the zero frequency group. The students’ quality of life, including physical functioning, role physical, bodily pain, vitality, social role functioning, role emotional, and mental health physical functioning, in the intermediate frequency group was moderately affected by patient initiated aggression.

From the data in Table 4, it is apparent that the scores of the eight dimensions of the SF-36 scale do not have strong correlations with patient initiated aggression frequencies. Additionally, correlation coefficients are not significantly different, which renders it unfeasible to assess the correlation between medical students’ quality of life and patient initiated aggression. In conclusion, patient initiated aggression did not affect the medical students’ quality of life.

There are several limitations in our study. First, a cross-sectional study cannot determine causality between patient initiated aggression and changes in medical students’ career planning or quality of life. Second, the self-reported nature of study data might introduce bias and errors. Finally, study participants practiced in a large hospital of a large city, which does not allow assessment of the situation in rural areas or small hospitals. The post hoc power is lower than 0.8, so this paper’s negative results is less credible. Large sample and longitudinal studies are warranted to confirm the findings and the issues of resources as a risk of violence.