Comparison of complaints to the intensive care units and those to the general wards: an analysis using the Healthcare Complaint Analysis Tool in an academic medical center in Taiwan

Enhancing responsiveness to the expectations of the population is a fundamental goal of the healthcare system [ 1]. The healthcare institutions had sought to achieve this goal by understanding the satisfaction of the patients, as has been advocated in the critical care setting [ 2– 5]. Information from the survey of satisfaction of intensive care unit (ICU) patients and families might infer issues for improvement of healthcare quality [ 6]. However, the information might be vague because respondents usually express a high level of satisfaction with the care [ 7] and the survey process may also deter them from expressing dissatisfaction due to concern about retribution by service providers [ 8, 9].

Recently, researchers and institutions have advocated the assessment of healthcare complaints as a complementary tool based on the views and perceptions of the patients and their families about the care and services. Healthcare complaints conventionally refer to an expression of grievance and dispute, typically written and communicated through a letter by a patient or their family, about the receipt of healthcare [ 10]. The complaints may request the correction of an ongoing problem or prevent its recurrence, while the complainers may believe they have valid information that the institution does not know or has failed to take sufficiently seriously. Similar to the evolution of surveys, the provision of complaints has also evolved to include multiple channels, such as e-mail and other electronic messaging services, and in-person contact with the patient relations office. This tool provides valuable information about people’s experience of the care and services received and their understanding of hospitals, patient safety, and healthcare quality [ 11, 12]. Since complaints are difficult to avoid in most clinical practice, reports recommend that the healthcare institutions include the analysis of complaints as a possible measurement of service quality [ 13– 17]. Nevertheless, few institutions have established a structured analysis and management system for complaints, while most of the complaints are managed by patient relations or risk management system. This practice is in contrast to the fact that most institutes have established patient safety monitor systems such as incident reporting systems [ 12]. Moreover, there have been a limited number of studies about the complaints related to intensive care medicine [ 18]. An Australian study reported that the complaint rate in the ICU setting, an estimated 5.9 per 1000 patients, was similar to that in the general wards; the ICU complaints focused mainly on treatment, communication, and access [ 19]. However, in comparison with questionnaires on patient satisfaction, previous methods used to analyze ICU complaints were not well-standardized and were without a widely accepted framework. Recently, researchers have developed complaint coding taxonomy [ 11]. The Healthcare Complaint Analysis Tool (HCAT) advocated by Gillespie et al. [ 20] provides a reliable tool for coding complaints and measuring the severity of complaints, to facilitate service monitoring and organizational learning and assessment of healthcare complaints as an indicator of poor service outcomes [ 20].

As healthcare complaints to ICUs may reflect preventable, ongoing, important problems related to the care of patients treated in the ICU, analyzing these complaints may be beneficial for the healthcare providers and the organization, to prioritize the domains that need timely communication and improvements in care. Furthermore, analyzing ICU complaints may provide insights not easily drawn from other sources, although a structured approach is also required. In this study, we applied the HCAT to analyze and compare healthcare complaints to the ICUs and the general wards of an academic teaching medical center, to understand the types of complaints and to investigate the factors associated with the severity of problems raised by the complaints.

We conducted this retrospective analysis at the National Taiwan University Hospital, Taipei, Taiwan. This study was approved by the institutional Research Ethics Committee C (#201706095 W) and was waived from the need for informed consent from the complainers or the patients.

The hospital, a 2300-bed, university-affiliated medical center with about 6400 full-time employees established the institutional Complaint Service Office (later renamed the Patient Relations Office) for management of healthcare complaints in 2008. The staff members of the office came from the Center for Quality Management of the hospital and served as the first-line correspondents for the complainers. They verified the complaint contents, collected pertinent information, contacted related workers of the hospital and the complainers and entered the data electronically into the institutional complaints management system. All of the verified complaints were filed and their contents were transcribed electronically into text files. The verbal contents from tape or video recordings of the complainers provided through non-written channels during the initial contact or in follow-up communications were also transcribed to text files. For this study, the investigators included all of the complaints to the ICUs during the study period and randomly selected twice the number of the complaints to the general wards during the same study period for comparison. The investigators retrieved relevant stored data from the office and coded the complaints through consensus for each complaint included in further analysis.

In this study, we estimated the rate and density of healthcare complaints to the ICUs. The complaint rate in the ICU was lower than the overall complaint rate for the inpatient setting in this hospital. We found that most of the complaints to the ICUs were issued from the family and surrogates, focused on the environmental issues when the problems were in the management domain and involved more problems focused on communication when the problems were in the relationship domain. Complaints to the ICUs were independently significantly associated with problems of higher severity among all complaints in the inpatient care setting. These findings provided insights into the opportunities for improving the care of patients in the ICU.

In addition, this study shows that the complaints provided information that may have been difficult to obtain from other sources such as the institutional Incident Reporting System. In this study, only 1.8% of the complaints had a coincident report as a patient safety event, indicating that the institution may not have received, at least in a timely manner, important information required to improve the quality of healthcare and services for the patients. The predominant categories in each domain of problems, such as quality versus safety, environment versus process, and listening/respect versus communication, suggest that complaints may be valuable as a complimentary tool to help understand the whole picture of healthcare and service quality provided by the institution.

Overall, the complaint rate related to the inpatient setting in our hospital was not as low reported in the literature [ 19]. A possible explanation is that our hospital has established multiple channels for the patients and their families to provide their complaints, which are all formally managed and filed by dedicated workers. In a study during 1997 to 2001, Taylor et al. demonstrated similar complaint rates in general wards and ICUs [ 19]; our study showed a lower rate of complaints to the ICUs than that to the general wards. Since we were unable to investigate the complaints before the year 2008, we did not know if there was a trend toward an increased complaint rate in the general wards than in the ICUs after 2001. However, the patients and families at our hospital applied multiple channels to issue complaints, like the feedback sheets, in addition to the traditional complaint process. It might be difficult to distinguish between comments with negative feedback and complaints, and other informal comments such as telephone calls, threats of lawsuits, and comments to caregivers commonly provided to the hospital as the sources for the collection of patient feedback [ 21]. The finding that the family and surrogates accounted for most of the contacts of the complaint is compatible with the premise that patients treated in the ICU setting might have difficulty in direct expression of undesirable experience in the care they received.

Our data may also reflect the healthcare context in Taiwan for general ward patients, in that there is almost no limitation on visits to the patient by family members or friends during the night shift, which is in contrast to the strict regulations for visiting patients in the ICU. The family members of the patients in the ICUs, with limited direct observation of the ongoing care and the inability of patients to express their needs and experience, may have had a quite different understanding of care compared to the general ward setting. We suggest that the differences in care setting and background context mean that it is necessary to understand the problems and concerns of care quality for the patients staying in an ICU.

Our finding that the patients made very few of the complaints that were issued suggests the need for a more effective channel to understand the unsatisfactory experience according to the ICU patients. Nevertheless, opinions from the family members might provide useful information on the healthcare system, as satisfaction among the families of critically ill patients has attracted increased concern as a primary indicator to evaluate care quality in the ICU. Previous reports about the surveys of ICU patients were mostly performed after surviving patients had left the ICU, but this could miss a significant proportion of non-survivors of the ICU. Survey of family satisfaction using a validated questionnaire has been suggested to assess satisfaction in such settings [ 2], but this type of survey might not render all of the answers to questions about the understanding of the experience by the patients and families. These tools might explore helpful information but usually request the responders to rate the experience under the pre-determined dimensions [ 22]. In a literature review of studies evaluating family satisfaction in the ICU, the mean response rate was 65.5%, and response rates greater than 70% were found only in 28.2% of the studies. Investigation of the complaints might, therefore, provide complementary understanding in this setting [ 12, 23]. As there was limited reporting of a structured analysis of complaints to the intensive care setting, we suggest that the hospitals should stress the analysis of healthcare complaints in this setting. Analyzing healthcare complaints systematically and recognizing situations and providers likely to generate complaints are critical steps in designing strategies for reducing complaints [ 23].

Patient safety was one of the main concerns raised by the patients and their family members through complaints and by the healthcare workers through the institutional incident reporting system. Nevertheless, the issues of safety raised by the complainers may be attributed to their understanding of patient safety. In addition, the complainer may have raised concerns related to conflicts about their expectations about care quality and the care environment. The classification of the complaints was derived using the HCAT; however, this does not mean that the patients were not at any risk of harm when there were complaints about environment and relationship issues. A broader definition of patient safety may include the additional risk of patient harm unrelated to the disease and treatment, and thus the issue of patient safety may be more wide-reaching than the “safety” category alone.

Our use of the structured classification and analysis of complaints to the intensive care setting as suggested by the HCAT may provide valuable insights into the clinical, managerial, and relationship problem raised by complaints in the intensive care setting. That the ICU complaints focused more on the environment than the institutional processes when the problem was in the management domain reflects the perception of the care environment, which might be very important for the family and surrogates. Also, the ICU complaints had more issues in the communication category than the ward complaints; this suggests that the complexity of care in the ICU might need better communication when the family faces the patient having a severe clinical condition and has important decisions to make. Indeed, the ICU environment was shown to be related to family satisfaction, which is reported to possibly be increased by 6% in the new ICU environment with noise-reduced, single rooms with daylight, adapted coloring and improved family facilities [ 24]. The report by Heyland et al. shows that the family is least satisfied with the waiting room atmosphere and frequency of physician communication [ 25]. Other themes for improvement reported in the literature included ICU atmosphere, amenities for visiting relatives, emotional support, consistency, clarity, and completeness of information, respect and compassion towards families, the inclusion of family, and support during decision-making processes [ 3, 4, 6, 25]. Despite the fact that family and relatives of the patients in ICU are expected to have both a receiving and participating role [ 26], families might also encounter unsupportive interactions, which might result from poor communication [ 27]. Our study findings provided further evidence of these observations, and stress the need to facilitate communication and interaction to relieve the anxiety and distress experienced by the families of the critically ill [ 27].

The use of patient satisfaction data to improve the process of care in the ICU has been recommended [ 28, 29]. As awareness and knowledge about healthcare has increased, demand from the public and individuals has surged. Likewise, complaints from patients could serve as a management tool for continuous quality improvement [ 17]. Therefore, healthcare should incorporate patient feedback as an integral part of quality improvement [ 17]; the patient’s perception of the quality of health services might come from the complaints [ 30]. Indeed, patients and their families might identify many issues as shown in our study that the existing healthcare quality frameworks do not cover [ 31]. While responsiveness remains a major domain of health system performance [ 32], the use of complaints to enhance the responsiveness might provide timely action for quality improvement. Despite difficulties remaining in the implementation of strategies to improve family satisfaction [ 33], we believe the analysis of healthcare complaints, including those to the ICUs, is of high value in quality improvement.

There are limitations to this study. First, this was a retrospective analysis based on the recorded opinions provided by the complainers, while the detailed information related to the problems during the interaction performed between the staff members of the complaint management and the complainers were not available; therefore, we cannot verify the problems, especially with the healthcare workers. Second, the original platforms for complaining and data entry of the complaint contents were not structurally designed to align with the HCAT; therefore, the study might miss some relevant issues as ideally recommended by HCAT. Nevertheless, as a significant proportion of complaints are still provided by the channels without direct contact or direct communication, the staff members of the complaint management or patient relations might not have the chance to go into details of the experience reported by the complainers. Third, some of the complaints were issued anonymously, therefore, we were unable to collect more detailed administrative data related to patient care. Indeed, there might be a trade-off process to obtain more opinions within the setting of anonymity, but with the loss of specificity on the understanding of the institutional care process and management issues. Fourth, the relatively small number of some types of problems found in the complaints might now be equivalent to the degree of patient satisfaction. We believe that the scarcity of healthcare complaints does not necessarily determine the fulfillment of high quality or complete satisfaction. Complaint management requires engaged personnel to understand the opinions of the complainers, similar to qualitative methods in the healthcare studies, especially in acute medicine such as in the ICU setting [ 34, 35]. To obtain more details and rationales for improvement, future prospective study of the combination of complaint interview and qualitative analysis is warranted.

In conclusion, complaints to the ICUs were associated with a higher severity of problems raised by the complaints to the institution. The analysis of healthcare complaints may provide valuable insights into the experience of care that the patients perceived in the intensive care setting, those usually issued from the family members of the patients, who might need more satisfactory communications and appreciation of a more comfortable care environment. The Healthcare Complaints Analysis Tool may render useful structure to classify and understand complaints in the intensive care setting, and thus warrants the establishment of this system for prospective collection of complaints data for the healthcare system.