Analysis of complaints lodged by patients attending Victorian hospitals, 1997–2001

Patient satisfaction — the subjective experiences of patients using the healthcare system — correlates with improved medical compliance,1 decreased utilisation of medical services,1 less malpractice litigation,1,2 and greater willingness to return to the healthcare provider.1-3 Accordingly, quantitative measurement of patient complaints is a comparative measure of service quality,4-6 and several authorities believe that quality-assurance measures should include patient satisfaction and an analysis of patient complaints.3,6-8

Complaints may arise from poor quality of service4,6 or unmet patient expectations.4,9 Some complaints appear minor, but many relate to more serious events and lead to remedial action or compensation. Analysis of the nature of complaints is important to identify problems and assist in their elimination.1,2,10

For quality-assurance purposes, individual hospitals may analyse and act on the complaints they receive. However, on any larger scale, the nature, frequency and outcomes of complaints have been poorly examined in Australia.

We have analysed data on patient complaints relating to a large sample of hospitals in the state of Victoria between 1997 and 2001, inclusive. We sought to identify subgroups of patients and hospital departments at higher risk of involvement in complaints, and to provide an evidence base for intervention strategies that aim to decrease patient complaint rates.

Complaints lodged at each participating hospital are categorised according to hospital department and the nature of the complaint using HCIP software. The major complaint categories are given in Box 1.

As there have been few studies of patient complaints at the state (or equivalent) level, many health professionals do not have comparison data on complaint rates.2 Our finding that female patients generate more complaints than male patients has been reported previously,4,12 but the reason is not known. Likewise, the higher complaint rate in public patients has not been explained. The lower complaint rate of non-Australian-born patients may relate to language difficulties confounding ethnic and cultural factors, and lack of familiarity with the healthcare system. However, contrary to our findings, Carrasquillo et al13 reported that non-English-speaking patients made more complaints.

Our overall complaint rate (1.42 complaints/1000 patients) is similar to that reported from one major Australian hospital (1.12 complaints/1000),7 but a higher rate was reported in a US hospital (5 complaints/1000).14 A previous report also found considerable variation in department complaint rates,10 but comparing department rates is difficult, as the nature of service provision varies considerably. For example, the clear difference in rates between the inpatient (aged care, general wards, intensive care) and outpatient/specialist departments may relate to “time at risk” or length of exposure to the hospital system.

Overall, the nature of the complaints is consistent with that reported by others,2,7,10,12,15 although billing and payment difficulties are more common in the United States.2,10 Complaints relating to communication were common, as also reported in other studies.2,6,7,10,12 This indicates a fundamental failure of staff to interact appropriately with patients. Furthermore, explanation, information provision, and resolution of misunderstandings contributed to a successful outcome for many patients, suggesting that communication problems may underpin most complaints lodged. In one respect, this could be encouraging, as relatively simple intervention strategies may have a profound impact on staff–patient interaction.

Access issues varied considerably. The finding that emergency departments were particularly vulnerable to complaints about treatment and admission delays is likely to be related to the problem of access block in Victoria’s emergency departments.16

It is encouraging that complaints relating to negligence, incompetence and wrong diagnosis were relatively uncommon. That emergency departments, intensive care units and operating theatres received the most complaints about treatment issues probably relates to the complexity of treatments provided in these departments. Although it is not known whether treatment issues did, indeed, reflect substandard treatment, relatively few treatment issues resulted in compensation payment.

Resolution of complaints was satisfactorily achieved in most cases, consistent with the findings of others.7,14 Importantly, an apology was acceptable in over a quarter of cases. However, apologies do not necessarily acknowledge incompetence or negligent treatment and may be given for any confusion or misunderstanding created, or lack of satisfaction with the service provided.6 Numerous authors have indicated that an apology given as soon as possible after a complaint may defuse the situation2,3,6,17 and reduce the time and resources required for final resolution.2,3

Australian Standards and resources on best practice and complaint management are available.18,19 Policy and protocol reviews are important in minimising adverse events.20 We found that only a small proportion of the complaints resulted in changes in policy or procedure.

Our study has several limitations. Firstly, the data analysed only represent patient complaints actually passed on to the hospital liaison officers and may under-represent the true complaint rates. Secondly, the incomplete nature of the databases may have introduced selection bias. Thirdly, the effect of age of the patient could not be analysed, as the HCIP database did not specifically record patient age.15 Fourthly, as the AIMS and VAED databases covered different time periods and utilised different software and data management procedures, there is a possibility of selection and measurement bias. Finally, despite HCIP classification guidelines, some complaint misclassification may have occurred.

We recommend the expanded use of complaint rates as quality assurance tools at the departmental and hospital level. High-quality data should be available for this purpose, and strategies to ensure that HCIP data are as complete and accurate as possible are indicated. Also, standardisation and streamlining of databases that record denominator data across all hospital departments will improve the accuracy of complaint rates. Finally, intervention strategies for high risk departments and patient subgroups are indicated to maximise patient satisfaction and minimise complaint rates.