Evaluation of Patients’ Perception of Safety in an Italian Hospital Using the PMOS-30 Questionnaire

Background: In our study, an Italian version of the PMOS-30 questionnaire was used to evaluate its feasibility and to improve health care quality in an Italian hospital. Methods: A cross-sectional study was conducted with 435 inpatients at a hospital in the Campania Region of Southern Italy using the PMOS-30 questionnaire and two other questions to assess patient feedback about the overall perception of safety. Results: The item “I was always treated with dignity and respect” showed the greatest percentage of agreement (agree/strongly agree = 89.2%; mean = 4.24). The least agreement was associated with the four “Staff Roles and Responsibilities” items (agree/strongly agree ranged from 31.5 to 40.0%; weighted mean = 2.84). All other 25 items had over 55.0% agreement, with 19 items over 70%. Moreover, 94.5% of the patients considered the safety of the ward sufficient/good/very good, and 92.8% did not notice situations that could cause harm to patients. Conclusion: Patient perception of safety was found to be satisfactory. The results were presented to the hospital decision makers for suggesting appropriate interventions. Our experience showed that the use of the PMOS-30 questionnaire may improve safety and health care quality in hospital settings through patient feedback.


Introduction
Patient safety is defined as the prevention of errors and adverse effects to patients associated with health care [1]. The importance of measuring patient safety in the health care setting has become widely recognized only in the past two decades. Therefore, before the 2000s, patient safety was not, with some exceptions [2], clearly incorporated into the dimensions of health care quality [3][4][5][6][7][8][9][10] and the agenda of health care organizations [11]. In 1984, Maxwell [3] identified the following dimensions of health care quality: access to service, relevance to need (for the whole community), effectiveness (for individual patients), equity (fairness), efficiency and economy, and social acceptability. This was followed by Donabedian's widely cited 1990 paper [4] that outlined seven dimensions of quality: efficacy, effectiveness, efficiency, optimality, acceptability, legitimacy, and equity. Lastly, in 1998, 11 dimensions and 21 sub-dimensions of health care service quality were identified; however, there were no items explicitly about patient safety [8]. Only with the new century has a safety culture permeated into the values and norms of health care organization members [12][13][14]. Now, patient safety has become an important element of care quality, first evaluated by experts and health professionals and ultimately including patients' feedback about their care [15][16][17].
In Italy, the Ministry of Health provided a variety of documents online about patient safety: recommendations for safety; reports on monitoring and analysis of sentinel events; reports on compliance from health care organizations with regional and national legislation; elaboration of many manuals and guidelines (about audit, root cause analysis, safety walkaround, safety in surgery, etc.) [18]. In recent years, several instruments and guidelines to improve patient safety have been developed using a patient-centred view [19]. Many of these evaluate patient feedback as a resource for improving patient safety in a hospital setting [20].
The Patient Measure of Safety (PMOS) is a recent structured questionnaire, designed to measure the patient's perception of safety [17,20]. Some authors of this paper are currently validating an Italian version of the PMOS-30 questionnaire [21].
In the present study, this Italian version of the PMOS-30 questionnaire was used in a hospital setting for evaluating its feasibility in an Italian hospital and for promoting the improvement of health care quality. This was a case study for extending the use of the PMOS-30 questionnaire in a wider Italian context.

Setting and Participants
An epidemiological cross-sectional study was conducted in one general hospital in the Campania Region, Southern Italy, to assess patient feedback on safety using the PMOS-30 questionnaire. All wards were involved, except the intensive/sub-intensive care, psychiatric, and COVID-19 isolation wards. The medical researcher visited the hospital five days a week. With appropriate scheduling, each ward was investigated every three to five days during the data collection period, and patients who had been hospitalised for at least three days were included.
Inclusion criteria were a minimum age of 18 years and Italian language speaking. Exclusion criteria included cognitive impairment, severe psychiatric disease, and end-stage disease. Participants were asked to sign an informed consent form.
The questionnaire was self-administered, and the patients were informed that participation was voluntary and that they could withdraw from the study at any time with no consequences. The medical researcher who delivered the questionnaire, a resident in Public Health, Epidemiology, and Hospital Organization, was independent of the hospital staff and was available to answer participants' questions about the questionnaire. He could not influence the patients, nor could he see what they were writing.
The time required to complete the questionnaire was about 15-20 min. After compilation, the completed questionnaires were immediately placed in a strictly private folder by the medical researcher; therefore, the privacy of patients was ensured, and the answers remained confidential. Patients were informed that all data collected would be analysed and aggregated and that their confidentiality would be strictly protected. The health care professionals of the wards were not informed about the content of the questionnaire.
Data were collected between August 2020 and November 2020. This study is a part of a research project conducted to validate an Italian version of the PMOS-30 questionnaire that will be published elsewhere [21]. In summary, the validation was carried out through confirmatory factor analysis and inter-item correlation. The English PMOS-30 questionnaire was translated into Italian and culturally adapted using standard forward-backward procedures performed by a multidisciplinary team [21].
Ethical approval for this study was obtained from the Ethics Committee of the University of Campania "Luigi Vanvitelli" (prot. N 0008664/i-2020).

Questionnaire
The questionnaire was divided into three sections. The first section contained an Italian version of the PMOS-30 questionnaire. The first version of the PMOS contained 44 items, but for the hospital setting, two shorter versions were created: PMOS-10 and PMOS-30 [22].
PMOS-30 included 30 items and 8 domains known to contribute to hospital safety: (1) communication and team working; (2) organisation and care planning; (3) access to resources; (4) ward type and layout; (5) information flow; (6) staff roles and responsibilities; (7) staff training; and (8) delays. All items were measured using a 5-point Likert scale (1-strongly disagree, 2-disagree, 3-neither disagree or agree, 4-agree, 5-strongly agree). There was also the option of "I prefer not to answer/I don't know". The second section consisted of two other items (not included in the PMOS-30) that investigated the patient's overall perception of safety through two direct questions and one open answer question: "How do you rate the safety of this ward?" (5-point Likert scale: 1-very bad, 2-bad, 3-sufficient, 4-good, 5-very good); "Have you noticed any events that could have caused harm to patients?" (yes, no); "If yes, describe" (open answer). The third section collected socio-demographic data and hospital characteristics: gender (male, female), age (18-40, 41-55, 56-70, > 70), education level (primary school, middle school, high school, degree), marital status (married, unmarried, other), employment (employed, housewife, retired), nationality (Italian, not-Italian), ward type (open answer), and days of hospitalization (continuous).

Statistical Analysis
For each item of the Italian version of the PMOS-30 questionnaire, mean and standard deviations were calculated, and for each domain, the weighted mean and Cronbach alpha were calculated. Scores for items with negative questions were reversed (items 5, 8, 9, 11, 12, 16, 17, 18, 19, and 20). Therefore, the percentage of agreement refers to the opposite meaning of the questions. "I prefer not to answer/I don't know", and blank responses were treated as missing values.
The patients' reported educational levels were standardised using the International Standard Classification of Education (ISCED 2011), which allows for cross-national comparisons of educational levels, and dichotomised into two groups: ISCED 0-2 (primary school, middle school) and ISCED 3-8 (high school, degree).
Several bivariate analyses were performed to determine if there were relationships between ward, days of hospitalization, and socio-demographic characteristics and some questionnaire items. Therefore, all the pertinent variables were dichotomized: ward type (medicine/surgery), days of hospitalization (3-5/ > 5), age (18-55/ > 55), marital status (married/other), employment (employed/housewife, retired), nationality (Italian/other), items of the PMOS-30 questionnaire (strongly disagree, disagree, neither agree or disagree/agree, strongly agree), and "How do you rate the safety of this ward?" (very bad, bad/sufficient, good, very good). Only associations <0.01 were considered statistically significant and presented in the results.
The sample size was estimated to be at least 400 subjects, assuming a 50% expected mean prevalence of "agree/strongly agree" in the PMOS-30 questionnaire, with precision of 5% and level of significance of 95%.

Socio-Demographic Characteristics
A total of 474 inpatients were approached, and 39 declined to participate; therefore, the response rate was 91.8%. Of the 435 participants, 55.2% were male (Table 1).
The respondents ranged in age from 18 to 90 years, and almost a third (30.1%) were over 70 years old. Education level was equally distributed between the two groups: primary and middle school (54.0%), high school and degree (46.0%). Half of the participants were not formally employed. Only 15 inpatients were not Italian (3.4%); this was lower than the percentages recently reported for non-Italians residing in Campania (4.6%) and in Italy (8.7%) [23].
The least agreement was found for the four items of domain 6, "Staff Roles and Responsibilities" (agree/strongly agree = 31.5-40.0%; weighted mean = 2.84). Question 8, "Staff didn't always know when a doctor changed my plan of care", had only 48.5% agreement (agree/strongly agree) and a high mean (3.49) because missing values were reported for 124 patients (28.5%). All the other domains and their 25 items showed agreement over 55.0%, with 19 items showing agreement over 70%. Conversely, except for domain 6, almost all (24/26) the other items did not exceed 17.0% of disagreement (disagree/strongly disagree). Many items garnered high level of the "I prefer not to answer/I don't know" response and blank responses (>50 in case of items 8, 11, 12, 21, 22, and 29). Weighted mean 3.54 Cronbach's alpha: N/A Agree = agree/strongly agree; Disagree = disagree/strongly disagree; Missing = "I prefer not answer/I don't know" and blank responses; N/A = not applicable; SD = standard deviation; (R) = reverse questions.
The results of the PMOS-30 questionnaire were then disaggregated for all the hospital and sociodemographic variables. In Table 3, only items with an association with p < 0.01 were included. Patients in the ISCED 0-2 group agreed more often than others with the eight items about safety (items 1, 2, 5, 6, 9, 17, 18, and 27). Conversely, they were less informed on three of the four items on "Staff Roles and Responsibilities" (items [23][24][25]. Furthermore, patients under 55 years of age had better knowledge about all four items on "Staff Roles and Responsibilities" (items [23][24][25][26]. In domains 1 and 6, Cronbach's alpha was >0.8, in domains 3 and 4 it was >0.6, and in domain 2 it was 0.525. Finally, the overall perception of the hospital's safety was evaluated using two other questions that were not included in the PMOS-30 questionnaire: "How do you rate the safety of this ward?" and "Have you noticed any events that could have caused harm to patients?". The majority (94.5%) of the patients considered the safety of the ward to be sufficient/good/very good, and 92.8% did not notice any situation that could cause harm to patients (Table 4). 27. Staff were always able to use the necessary equipment 0.004 (R) = reverse agreement (i.e., ISCED 3-8 vs. ISCED 0-2); * strongly disagree, disagree, neither agree or disagree/agree-strongly agree. Among the remaining 31 patients who reported potential harms, more than half of the reports were related to the ward staff (incompetence; poor attention, poor communication, and poor information to patients; delays in health care). Other answers concerned patient falls, structural deficiencies, excessive noise, and poor cleaning. There were no differences in the answers given by patients with different sociodemographic characteristics.

Discussion
In 2018, the OECD stated that the true extent of safety and harm across all health care settings is still a black box. Therefore, measuring safety is the starting point for improving patient safety because without measurement, actions to drive improvement are impossible [13].
We assessed patient feedback on safety using an Italian version of the PMOS-30 questionnaire. To date, the PMOS questionnaire has been validated in the Australian [24] and Persian settings [25]; this is the first time that the PMOS-30 questionnaire has been used in an Italian hospital setting.
Overall, the results of the PMOS-30 questionnaire showed satisfactory patient feedback on safety, considering that the lowest mean was 2.7 (item 24) and that all means were above 3.4, except in the "Staff Roles and Responsibilities" domain.
Patients agreed most often with the item "I was always treated with dignity and respect". This result is in agreement with that obtained by Taylor et al. [24]. On the other hand, patients' knowledge of "Staff Roles and Responsibilities" was the lowest. This result stems from a characteristic of many Italian health care professionals, an avoidance of being recognized; therefore, there is a low propensity to communicate one's name to patients. For this reason, the Italian legislation decreed, in 1995, that patients must be informed about the identity of the health care staff, who must attach an identification tag to their gowns [26].
It is not surprising that there was a high number of "I prefer not to answer/I don't know" and blank responses because there were many questions that patients might not have been able to answer. For example, the two items with the highest numbers of missing values ("I prefer not to answer/I don't know" and blank responses) were "Staff didn't always know when a doctor changed my plan of care" (n = 124) and "Information about me that my health care team needed was always available (e.g., drug charts, medical notes, test results)" (n = 98). These missing data probably resulted from poor communication between health care professionals and patients, so patients might not have asked about these interactions, or not have felt empowered to ask about these interactions. This aspect of communication is another topic often emphasized in continuing education courses for health care professionals. Better communication between health care professionals and patients leads to better outcomes with less cost and work efforts [27,28].
The overall perception of safety (determined using two questions not included in the PMOS-30 questionnaire) was satisfactory. Only 24 patients judged safety as bad/very bad. Only 31 identified events that could have caused harms, more than half of which concerned ward staff behaviours, among which poor attention, communication, and information to patients prevailed. These data align with the previous results that highlighted the issue of poor communication between health care professionals and patients [28].
The results of our study were presented to the hospital decision makers with the aim of suggesting interventions to improve safety in those areas where the less satisfactory results were obtained. The main intervention was to improve the relationship and communication between health care professionals and patients. The decision makers were also presented with a list of the 31 potential harms reported by patients to allow selective interventions in individual wards. The hospital health manager also shared the results of the study with the heads of the various selected wards to make it easier for the heads to recognize weaknesses and take effective corrective action. Therefore, with this instrument, targeted improvements have been made in individual wards to increase overall patient safety.
The study had the following limitations. This was the first experience with the PMOS-30 questionnaire within a single Italian hospital. While satisfactory results for improving patients' safety have been achieved, the instrument needs to be applied in a wider context to better assess its feasibility. Moreover, questionnaire administration to inpatients can create anxiety in the patients about reporting undesirable results because those may become known by the staff. Consequently, this could reduce reporting of negative answers. Therefore, many measures were put in place to reduce patient anxiety about this occurrence, as described in the Methods section. However, a residual fear cannot be ruled out. Regarding the question out of the PMOS-30 questionnaire ("Have you noticed any events that could have caused harm to patients?"), we have not informed the participants which events could lead to harm. Therefore, the answers reporting potential harm to patients might have been underestimated.

Conclusions
We believe that, if the hospital decision makers are involved, the use of the PMOS-30 questionnaire might improve safety and health care quality in hospital settings through patient feedback. Future research studies with the routine use of the PMOS-30 questionnaire in a hospital setting might highlight gaps and deficiencies in individual wards and allow focused improvement. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.

Conflicts of Interest:
The authors declare no conflict of interest.