The growing complexity of primary health care delivery presents significant challenges for GPs [4]. Given the important, and under-utilised insights of patients into the care process, the aim of this study was to assess patients’ perceptions of safety climate in Irish general practice, determine whether perceptions differed according to various patient risk factors, and explore whether open-ended responses of patients enhanced our understanding of patient safety information.

Overall, participants reported positive perceptions of safety across all subscales of the PPS-GP. Responses to global safety items were also positive, with participants scoring a median response of 9.0 (IQR = 2.0) for overall practice safety rating, on a scale from 0 to 10, and 84.9% ‘likely’ or ‘extremely likely’ to recommend the practice to friends and family. Few other studies have administered patient-report questionnaires to assess safety in the primary care setting [15], which limits our ability to draw extensive comparisons with existing research. Of those in existence, the PREOS-PC conducted in England [30] found that patients had generally positive perceptions of the safety of care provided in general practice, with a mean score of 84.6 out of a potential 100, and 91% agreeing that their HCPs were trustworthy. In a study of patient perceptions of the safety of primary chronic care in Finland [7], 68% either agreed or strongly agreed that they received safe care at home. In a patient-report measure of patient experience of patient-centred medical homes in the US [31], 63% gave positive ratings to their clinic on confidence in quality/safety. It therefore appears that patient perceptions of safety in Irish general practice are more positive than those from other international studies.

The level of favourable views in relation to the ‘Staff stress and workload’ subscale is perhaps surprising, given that a 2015 study of GPs in Ireland reported that 74% rated their stress levels as ‘high’ or ‘very high’ [32]. Further, in a survey of burnout amongst a sample of GPs working in Ireland [33], 52.7% reported high levels of emotional exhaustion, and 6.6% fulfilled the criteria for burnout. However, O’Dea et al. [33] acknowledge that despite high stress levels, Irish GPs continue to derive satisfaction from their work as compared to their international counterparts, and patients are less likely to receive substandard care. This may partly explain why staff workload does not appear to translate to poor patient perceptions of safety. Although this is a relatively positive finding, it is important to also consider the possibility that patients may not actually ‘see’ such system issues. Indeed, a number of qualitative comments which were excluded from the content analysis suggested this (e.g., ‘It’s difficult to answer some of the questions in the survey as its often not possible for me to know. For example, my doctor is stressed- how would I know unless they told me so. They could be extremely stressed but either hiding it well or not even aware if their own stress levels’). Future research may therefore consider incorporating multiple perspectives of safety climate (e.g., patient and HCP perceptions) in order to obtain an accurate, full picture on systems factors.

Regarding patient risk factors, older age and being of Irish nationality were the only predictors that were significantly associated with positive SC perceptions. Similarly, De Voe et al. [34] report that patients aged over 65 years had positive perceptions of communication with healthcare providers. These patients felt that providers listened to them, showed respect for what they had to say, and spent enough time with them as compared to those aged 18–64. Such perceptions are in spite of findings that older patients are at a greater risk of experiencing a PSI in primary care [35], most commonly related to medication-related incidents, communication-related incidents, and clinical decision-related incidents [36]. Nevertheless, our study demonstrates that older patients feel safer receiving care in Irish general practice- despite their increased risk profile and greater susceptibility to PSIs. We would have also expected participants reporting the presence of certain risk factors such as multimorbidity and polypharmacy would be associated with lower perceptions of safety, given that their complex health profile places them at an increased risk of safety issues [18], and has been associated with a higher occurrence of PSIs [5, 6]; however, poorer SC perceptions were not evident amongst these participants. That non-Irish respondents to our survey felt less safe receiving care than Irish people warrants further exploration to ensure culturally sensitive, safe primary healthcare delivery by targeting language barriers, training needs, and developing guidelines for effective cross-cultural communication [37]; particularly in light of health equity research finding that people of colour are more likely to experience patient safety events [38].

The majority of free-text responses were related to communication and feedback, which is unsurprising given that effective communication has been consistently identified by patients [39, 40] as a key contributor to PSIs. Deficits in relation to access and timeliness were also frequently identified, such as difficulty in obtaining an appointment, which has been cited as a driver of safety problems in other studies [30]. It is, however, surprising that over two-thirds of the responses related to ‘poor’ SC practices, given that the overall SC perceptions were so positive. This finding is similar to research conducted in a hospital setting, whereby patients cited widespread criticism of the hospitalisation experience in response to an open-ended question, despite reporting high satisfaction scores in response to closed-ended questions [41]. The use of an open-ended option in the current study therefore allowed for the exploration of divergent responses to closed-ended questions, and raised issues that would likely have been less noticeable otherwise. This highlights the advantages of using qualitative methods to derive data that provide a deeper and more nuanced understanding [41] of the care experience than collecting quantitative data alone. However, given the disadvantages associated with the use of one format alone (e.g., item non-response to open-ended questions due to time burden [42]) we would suggest that future research combines the use of both qualitative and quantitative methodology when exploring patient-reported safety perspectives. Additionally, these findings emphasise the valuable role that patients can play in identifying poor practices, thus providing information that can be used by GPs to inform safety improvements. This is particularly useful in light of research suggesting reporting that primary care physicians have cited difficulty in understanding how best to measure and improve patient safety in their practices [17].

Future research

Our finding that patients predominantly identified ‘poor’ SC practices, despite reporting generally positive SC perceptions suggest the need to explore isolated incidents of safety in general practice in greater detail. Previous research has found that even patients with generally positive perceptions of care could recall at least one safety incident they had witnessed previously [43]. Similarly, Ricci-Cabello et al. [30] found that despite participants reporting that providers took adequate measures to ensure safe healthcare delivery, 45% reported experiencing at least one safety problem in the previous 12 months. While the examination of specific incidents was outside the scope of our novel measure, this suggests that despite favourable general perceptions of SC, there may exist a need for future research to further explore the occurrences of isolated incidents of harm.

It has been suggested that ‘a single measure of safety is a fantasy’ [11] and given our findings, the gathering of patient-reported safety information is no exception. There are various purposes, strengths and weaknesses associated with the use of each patient safety measure, which must be considered as complementing each other by providing different levels of qualitative and quantitative information [44]. Therefore, the triangulation of various data collection methods has been recommended to obtain a full view [44, 45] of the safety experience, and ought to be applied to the general practice setting.

Although participants had generally positive perceptions in relation to ‘Patient knowledge and accountability’, just over half reported knowing how to report issues with their care. Similarly, Ricci-Cabello et al. [30] reported low levels of patient activation, with the majority of participants reporting that they ‘never’ or ‘rarely’ raised a concern when they thought something was wrong. In a study of patient complaints, O’Dowd et al. [46] cite a lack of knowledge of the complaints process as a potential reason for patients not complaining [47]. Efforts should be made to ensure that patients are aware of the processes involved in reporting issues with their care, either at a practice or a national level.

Strengths and limitations

There are a number of limitations that should be considered when interpreting the results of the current study. First, we were unable to calculate a response rate for the online version of the questionnaire, as the recruitment strategy used did not make it possible to collect data on how many patients were invited to complete it. Further, it is reasonable to suggest that it may be lower than the paper version of the survey, given that web response rates have been consistently found to be lower than rates achieved using traditional data collection methods in public health research [48]. Despite this, our response rate is considerably higher than other patient-report safety measures conducted across primary [30] and secondary care [49, 50] settings.

Second, our study sample may not be representative of the population as a whole. Participants self-selected into the study, which may have imparted a self-selection bias, whereby certain types of participants were more likely to participate (e.g., those who were more motivated and more positive [51]). Further, 75% of respondents were female; although Irish data has suggested that women use GP services more frequently than men [52], our figure is disproportionate. Despite contacting a number of chronic illness support groups to share study information, only 14% of participants reported multimorbidity. It would be expected that this would be higher in a representative sample, given that approximately 27% of Irish adults report the presence of at least one long-standing illness or health problem [52]. The majority of respondents were also of Irish nationality, which may be partly explained by the PPS-GP being solely administered in English. Therefore, future research in this area ought to give further consideration to targeted recruitment strategies to capture the safety perceptions of those less well represented populations (i.e., male and multimorbid patients), with a particular emphasis on engaging non-Irish participants (e.g., by translating the measure into different languages), particularly in light of our findings that non-Irish respondents have poorer SC perceptions.

Third, although some significant regression coefficients were observed, suggesting that age and nationality are related to SC perceptions, a small portion of the variance in SC was explained by our set of predictors. This would suggest that our regression models provided poor fit. However, it has been reported that low variance in regression models have been consistently found in previous patient-report studies of healthcare [41] and this is relatively common in social sciences research.

Finally, some, but not all, of the participants were recruited in the midst of the COVID-19 pandemic. Although there is no known research to date specifically on patient-reported perceptions of safety in primary care during COVID-19, as we have acknowledged previously [16], some studies have reported that patient satisfaction has been found to be higher in the COVID-19 period than in the period immediately before [53]. It is therefore possible that there was an artefact from the pandemic (e.g., positive skewness of item responses), given the established links between patient safety perceptions and patient satisfaction [43].

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.


This article is autogenerated using RSS feeds and has not been created or edited by OA JF.

Click here for Source link (