Secondary Logo

Journal Logo

An Analysis of Risk Factors for Patient Complaints About Ambulatory Anesthesiology Care

Kynes, J. Matthew MD*; Schildcrout, Jonathan S. PhD†‡; Hickson, Gerald B. MD§; Pichert, James W. PhD§; Han, Xue MPH; Ehrenfeld, Jesse M. MD, MPH*; Westlake, Margaret W. MLS§; Catron, Tom PhD§; Jacques, Paul St. MD*

doi: 10.1213/ANE.0b013e31827aef83
Economics, Education, and Policy: Research Report
Free
SDC

BACKGROUND: Anesthesiology groups continually seek data sources and evaluation metrics for ongoing professional practice evaluation, credentialing, and other quality initiatives. The analysis of patient complaints associated with physicians has been previously shown to be a marker for patient dissatisfaction and a predictor of malpractice claims. Additionally, previous studies in other specialties have revealed a nonuniform distribution of complaints among professionals. In this study, we describe the distribution of complaints among anesthesia providers and identify factors associated with complaint risk in pediatric and adult populations.

METHODS: We performed an analysis of a complaint database for an academic medical center. Complaints were recorded as comments during postoperative telephone calls to ambulatory surgery patients regarding the quality of their anesthesiology care. Calls between July 1, 2006 and June 30, 2010 were included. Risk factors were grouped into 3 categories: patient demographics, procedural, and provider characteristics.

RESULTS: A total of 22,871 calls placed on behalf of 120 anesthesiologists were evaluated, of which 307 yielded a complaint. There was no evidence of provider-to-provider heterogeneity in complaint risk in the pediatric population. In the adult population, an unadjusted test for the random intercept variance component in the mixed effects model pointed toward significant heterogeneity (P = 0.01); however, after adjusting for a prespecified set of risk factors, provider-to-provider heterogeneity was no longer observed (P = 0.20). Several risk factors exhibited evidence for complaint risk. In the pediatric patient model, risk factors associated with complaint risk included a 10-year change in age, the use of general anesthesia (versus not), and a 1-hour change in the actual minus scheduled start times. Odds ratios were 1.47 (95% confidence interval (CI), 1.04–2.08), 0.22 (95% CI, 0.07–0.62), and 1.27 (95% CI, 1.10–1.47), respectively. In the adult patient model, risk factors associated with complaint risk included male gender, general anesthesia, a 10-year change in provider experience, and speaking with the patient (rather than a family member). Odd ratios were 0.66 (95% CI, 0.47–0.92), 0.67 (95% CI, 0.47–0.95), 1.18 (95% CI, 1.01–1.38), and 1.96 (95% CI, 1.17–3.29), respectively.

CONCLUSIONS: There was apparent evidence in adult patients to suggest heterogeneity in provider risk for a patient complaint. However, once patient, procedural, and provider factors were acknowledged in analyses, such evidence for heterogeneity is diminished substantially. Further study into how and why these factors are associated with greater complaint risk may reveal potential interventions to decrease complaints.

Published ahead of print February 5, 2013 Supplemental Digital Content is available in the text.

From the *Department of Anesthesiology, Vanderbilt University School of Medicine, Nashville, Tennessee; Departments of Biostatistics and Anesthesiology, Vanderbilt University, Nashville, Tennessee; and §Center for Patient and Professional Advocacy, Vanderbilt University School of Medicine, Nashville, Tennessee.

Accepted for publication October 1, 2012

Published ahead of print February 5, 2013

Funding: Internal funding.

See Disclosures at end of article for Author Conflicts of Interest.

Reprints will not be available from the authors.

Address correspondence to Paul St. Jacques, MD, Department of Anesthesiology, Vanderbilt University School of Medicine, 1301 Medical Center Drive, 4648 The Vanderbilt Clinic, Nashville, TN 37232-5614. Address e-mail to paul.stjacques@vanderbilt.edu.

Patient safety and satisfaction are of paramount importance to the practice of anesthesiology. Underscoring this importance are regulatory efforts to examine physician professional practice performance in credentialing, reimbursement, and other evaluation metrics. Efforts, such as patient complaints tracking, are mandated by the Centers for Medicare and Medicaid Services.1 Aggregated complaint data have been shown to serve as an important measure of quality care, and a thorough analysis of complaint content has been used to promote performance and quality.2–8 Studies in several medical specialties have shown that patients’ complaints are associated with risk management activity and medical liability claims.9–17 Additionally, the literature regarding risk mitigation has stressed the importance of the assessment of patient satisfaction in preoperative18,19 and postoperative20–22 settings.

The literature presents a variety of studies on patient-based outcome measures related to ambulatory care.23–25 Early studies of pre- and postanesthesia care include quantitative and qualitative data related to outcomes and symptoms, and many also report on patient satisfaction.20,23,24,26–29 These studies include patient satisfaction data aggregated by complaint type, yet none report on the distribution of patient complaints associated with specific anesthesia providers, patient factors, or procedural factors. In other fields, identifying providers associated with patient complaints is a proven methodology for reducing malpractice risk and improving patients’ perceptions of quality of care.3,8,11,12

Although the distribution of complaints among anesthesia providers has not been previously reported, studies have shown that patient complaints are not evenly distributed among physicians either between11,12 or within specialties.9,16,30 These studies used unsolicited complaint reports initiated by patients or families and were collected in an organization’s Patient Relations office. In these studies, approximately 50% of physician-related complaints were associated with small proportions (8%–14%) of the physicians.16 Some explanatory factors for this nonuniform distribution have been proposed, but complaint risk factors have yet to be fully elucidated or examined in anesthesiology.

In this study, we examined the distribution of patient complaints among anesthesia providers and sought to identify patient, physician, and procedural risk factors for these complaints. Identifying determinants of this risk for complaints could provide a window for intervention to improve patients’ health care experiences and decrease complaint frequency.

Back to Top | Article Outline

METHODS

Study Setting

Data for this study were collected from postprocedure telephone surveys of ambulatory surgical patients at an academic medical system. This center provides anesthesia services for approximately 60,000 adult and pediatric patients per year at >70 anesthetizing locations throughout the medical center. Ambulatory patients, those who enter and leave the hospital on the same day as their surgery or procedure, comprise approximately 21,000 cases annually. The center employs a large academic anesthesiology group, all of whom were eligible for inclusion in this study. Case coverage primarily follows the anesthesia care team model with certified registered nurse anesthetists or resident anesthesia providers present for almost all cases.

Back to Top | Article Outline

Description of Callback Database and Procedure

This study was reviewed and approved by the IRB. It was designed as a retrospective descriptive analysis of ambulatory surgery patient complaints. The period studied was from July 1, 2006 through June 30, 2010. Data were queried from a perioperative electronic database (VPIMS, Acuitec, LLC, Birmingham, AL). The database includes a specifically designed component used to record patients’ responses to a phone survey on postsurgical outcomes, satisfaction, and their care experiences.

As part of the routine postoperative process, calls to patients were initiated from a nonselective daily list of patient names and phone numbers arranged by procedure-related location. The patients were listed alphabetically by last name. On alternate days, the caller began at either the top or bottom of the list and placed as many calls as could be completed during their shift. The caller attempted to reach patients via telephone within 24–48 hours after surgery, made 1 attempt to reach as many listed patients as possible during their shifts, and were blinded to the name of the anesthesia provider at the time of the phone call and data entry. Surveys were counted as having been completed if the caller talked with the patient or alternately a family member who claimed familiarity with the care the patient received and who agreed to respond to the questions.

If successfully contacted, the caller read through a script of 19 questions and manually typed patients’ or responders’ comments into the database during the call, as each question was answered. As part of the script, patients were asked to rate satisfaction with their anesthesia care as excellent, good, fair, or poor. For responses other than excellent, patients were asked what aspect of their care was less than excellent. The text of the response, including direct quotes, was recorded into the database for subsequent complaint analysis.

Response data were electronically transferred to the Vanderbilt Center for Patient and Professional Advocacy for coding and analysis. The Vanderbilt Center for Patient and Professional Advocacy used the Patient Advocacy Reporting System®,8 a validated method to analyze complaint data and code distinct complaints identified within each patient response. Only complaints regarding anesthesia care were evaluated, and more than 1 complaint could be coded for each patient response. Complaints were coded into 6 categories that have high inter-rater and test–retest reliability.3,8,12,31 The 6 categories included complaints related to care and treatment (e.g., I was given nothing to control my nausea), (58%); communication (e.g., used terms that I did not understand), (25%); access and availability (e.g., Dr. __ did not visit following my surgery), (4%); concern for patient and family (e.g., Dr. __ was rude and indifferent), (14%); safety of environment (0%); and money or payment issues (0%). A complaint type summary for each physician was created based on the distribution of complaints among the 6 complaint categories.

Back to Top | Article Outline

Data Management and Exclusion Criteria

Over the course of the observation period (July 1, 2006 to June 30, 2010), 1 primary caller was used. However, from June 2008 through December 2008, while the primary caller was on leave, a temporary replacement was responsible for delivering the telephone surveys. During that time period, complaint rates decreased dramatically (by >80%). The research team concluded that the 2 callers were not exchangeable, and furthermore, that the replacement caller did not adequately follow protocol and did not establish the rapport necessary to ascertain reliable survey data. Data from the second caller were excluded from the analysis.

One important variable for this analysis involved whether or not the caller spoke with the patient or to a family member. Among the observed calls, this variable was missing on approximately 6% of cases. We only included data from cases in which these data were available. Furthermore, in approximately 97% of cases when the patient was a minor, the caller spoke to a family member as opposed to the patient. Among the 3% of calls where we spoke with the (minor) patient, no complaints were recorded. These calls were removed from the analysis since regression modeling requires variation to estimate model parameters.

We limited all analyses to procedures (determined by primary CPT code in the perioperative database) that were performed on at least 3 occasions and by anesthesiologists for whom at least 3 phone calls were made. Five anesthesiologists were removed from analyses due to inadequate call volume. After these exclusions, we reported results based on 22,871 survey calls made on behalf of 120 physicians that pertained to 901 unique procedures.

Back to Top | Article Outline

Statistical Analysis of Risk Factors

For pediatric and adult population models, risk factors were grouped into 3 categories that capture distinct features of each surgical encounter: (1) patient demographics, (2) procedural factors, and (3) provider characteristics. Patient factors included patient age, gender, and race (non-Caucasian versus Caucasian race). Procedural factors included procedure based on primary surgical CPT code, anesthesia type (general, regional, and monitored anesthesia care), ASA status, and the time from the scheduled to actual surgery start time. Provider features included provider race, age, gender, years in practice, and the number of days the provider worked during the month each surgery was performed. The last 2 variables were intended to capture factors related to experience and workload. Two exogenous factors that were also examined were the date of the surgery (to capture secular/temporal trends) and whether or not the caller spoke to the patient or to a family member. The latter variable was only examined in the adult population model because all calls included in the analysis of minors were received by patient family members.

For the pediatric and adult patient models, the primary outcome variable was binary, and its value was set to 1 if, during the follow-up telephone call, the patient or guardian registered a complaint regarding anesthesiology care. The outcome value was set to zero otherwise. To examine risk factors for patient complaints and the extent of provider-to-provider heterogeneity in complaint risk, logistic-normal generalized linear mixed effects models with provider-specific and procedure-specific random intercepts were used (i.e., non-nested, crossed random effects). Both a null model that included no risk factors and a risk factor–adjusted model were used. The use of both models allows for examination of apparent provider-to-provider heterogeneity (by estimating and testing variance components in the null model) and a more realistic estimate of heterogeneity above and beyond categories of risk factors (by estimating and testing variance components in the risk factor–adjusted model).

To formally examine provider-to-provider heterogeneity, variance components were estimated and tested. The variance component associated with the random intercept for anesthesiologists is a measure of the extent of heterogeneity in complaint risk among providers. As has been discussed elsewhere,32,33 to test the null hypothesis that this heterogeneity factor (i.e., the variance component) is zero is challenging because variances can only be positive. Thus, the null hypothesis lies on the boundary of the parameter space. The test was conducted using a permutation test31 based on 2000 random reshufflings of the provider identifiers.

Risk factor effect estimates are reported with odds ratios (ORs) and 95% confidence intervals (CIs). For the continuous covariates, likelihood ratio tests for nonlinear functional forms of effects using restricted cubic splines were performed. Those covariates evidencing a nonlinear relationship with the log odds of a complaint are displayed graphically and those for which there was no such evidence are reported in tabular form along with categorical variables. We do not provide CIs for variance components because they are of questionable interpretability particularly when they are close to the zero.

It is worth highlighting that there were fewer complaints in the analysis of the pediatric population (121 complaints) than in the adult population (186 complaints), and we had some concerns about model overfitting in both models. Though we examined the impact of regional anesthesia in adults, we do not do so in children due to concerns regarding overfitting. Additionally, because monitored anesthesia care is highly negatively associated with general anesthesia in both populations, it was not included in regression models. Calibration plots were used to visually inspect the prespecified model fits, and there was little evidence of model overfitting.

To examine the practical importance of the excess risk estimates provided from analyses, we combined the mixed effects model results with the distributions of the risk factors that were observed to be significantly associated with complaint risk to derive an estimate of the number of complaints that could have been avoided in the absence of excess risk. We call these the theoretical number of preventable complaints. For categorical variables, this involved examining the difference between overall estimated number of complaints in the observed sample and what would have occurred had risk in the high-risk stratum been decreased to that of the low-risk stratum. That is, we estimate the quantity: N × RD × p where N is the number of patients, RD is a risk difference between the high- and low-risk strata, and p is the proportion of patients in the high-risk stratum. For continuous covariates, the calculation involved estimating the difference between overall estimated number of complaints in the observed sample and what would have occurred had those above the median risk been reduced to the median risk.

Back to Top | Article Outline

RESULTS

Tables 1 and 2 provide summaries regarding provider characteristics and call completion rates, and Figure 1 shows complaint rates per provider and per unique procedure to display heterogeneity in complaint risk without acknowledgment of risk factors surrounding the encounter. Table 3 describes the distributions of the risk factors. Of the 22,871 calls included in analyses, 10,194 pertained to the pediatric sample and 12,677 pertained to the adult sample. Overall, 307 (1.34%) patient complaints were reported. In the pediatric population, the median interdecile range age was 4 [0.7, 14] years and 6 [1, 15] years in calls where a complaint was not and was reported, respectively. The median [interdecile range] actual minus scheduled start time was 4 [−56, 62] minutes and 16 [−27, 97] minutes in calls without and with a complaint, respectively. In adult patients, speaking to the patient, female gender, and nonutilization of general anesthesia each showed some evidence of increased risk associated with a complaint, although we caution against the overinterpretation of these unadjusted comparisons that are subject to confounding. Results for the primary analyses that seek to examine risk factors for a patient complaint are shown in Table 4 and Figure 2.

Table 1

Table 1

Table 2

Table 2

Table 3

Table 3

Table 4

Table 4

Figure 1

Figure 1

Figure 2

Figure 2

Back to Top | Article Outline

Pediatric Models

Patient age, general anesthesia usage, and actual minus scheduled start times were each associated with complaint risk. The ORs associated with a 10-year change in age, the use of general anesthesia versus not, and a 1-hour change in the actual minus scheduled start times were 1.47 (95% CI, 1.04–2.08), 0.22 (95% CI, 0.07–0.62), and 1.27 (95% CI, 1.10–1.47), respectively. The theoretical number of preventable complaints associated with higher age, nonuse of general anesthesia, and higher actual minus scheduled start times were estimated to be 16 [2, 34], 3 [1, 9], and 12 [4, 22], respectively. Normalizing these data across the total cases analyzed yields a rate of 0.16%, 0.03%, and 0.12% of all cases, respectively. The estimated random effect SD due to provider was very near to zero, indicating little or no evidence of provider-to-provider heterogeneity in risk of complaints.

Back to Top | Article Outline

Adult Models

Patient gender, general anesthesia, provider years in practice, and whether or not the caller spoke with the patient were each associated with complaint risk. The OR for male patient gender (versus female), for general anesthesia (versus not), for a 10-year change in provider experience, and for speaking with the patient (versus a family member) were 0.66 (95% CI, 0.47–0.92), 0.67 (95% CI, 0.47–0.95), 1.18 (95% CI, 1.01–1.38), and 1.96 (95% CI, 1.17–3.29), respectively. The theoretical number of preventable complaints associated with female gender, nonuse of general anesthesia, and with higher provider years of experience were estimated to be 45 [8, 78], 24 [3, 46], and 16 [1, 34]. Normalizing these data across all cases analyzed yields a rate of 0.35%, 0.19%, and 0.12% of all cases, respectively.

Even though there appeared to be provider-to-provider heterogeneity in the null model where the estimated random effect SD was 0.35 (P = 0.01), once risk factors were added to the regression, the evidence of such heterogeneity was reduced substantially. The random effect SD was estimated to be 0.18 (P = 0.20) in the risk factor model. To interpret these variance components further, if we were to compare the median to the 95% percentile performing anesthesiologist, the estimated variance components imply an OR of 2.01 (median to 95% percentile) in the null model and an OR of 1.43 in the risk factor model. Thus, the risk factors appear to have “explained” a large percentage of the apparent provider-to-provider heterogeneity observed in the null model. We also note that we conducted an identical analysis but only included calls made to the patient, and the observed heterogeneity in the risk factor model was approximately 0.10 (rather than 0.18 in this model).

Back to Top | Article Outline

Secular Trends

Figure 2 shows secular trends in complaint rates estimated from the risk factors models in the pediatric and adult populations. Over the course of the study period, patient complaint rates decreased until approximately January 2009 at which point in time complaint rates began to increase.

Back to Top | Article Outline

DISCUSSION

Similar to previous studies, our study reveals that in specified subgroups in an ambulatory surgery setting, younger patient age,19,26 female gender,19 case delay,28 and regional anesthesia,28 or IV sedation were associated with a higher risk of complaints. Some of the risk factors identified appear to have clear associations for driving complaints. Case delay, for example, may create frustration by prolonging a preoperative period of anxiety and fasting. Other risk factors, including provider characteristics and patient demographic factors reflecting willingness to complain, warrant further investigation.34 The distribution of complaint types observed in this study was similar to those obtained in a study of complaints associated with urologists.16 In both studies, complaints about care and treatment predominated, and complaints about communication accounted for approximately 25% of the totals. Types of complaints may differ across specialties, but any particular physician’s complaint type distribution may reflect important opportunities for the physician to identify and address any patterns in those complaints. The temporal trend in complaint rates was an interesting finding that might reflect national trends in complaint increases since the economic downturn of 2008. The Better Business Bureau, for example, has noted a trend of consumer complaint increases of 7%, 10%, and 10% in 2008, 2009, and 2010, respectively.a

Data collection methods play an important role in survey results of patient satisfaction. Regardless of method, overall patient satisfaction with anesthesia care is generally high.28,29 More positive ratings of patient satisfaction are recorded during face-to-face interviews, second from phone interviews, and third from mail surveys.35 The higher satisfaction scores may be attributed to the respondent’s desire not to offend or displease the surveyor, to minimize negative repercussions in their future health care.24,28,29,35 Because patients are reticent to criticize, recorded patient complaints become important data. This study analyzed patient satisfaction data recorded in a narrative text field from postoperative phone call system. Despite the limitations of this system, significant trends were still established over the study period.

Reducing the drivers of complaints directed toward high complaint risk categories has the potential to increase patient satisfaction and care quality. While the contribution of other physician-specific factors appears to be modest, information gained from physicians who have low numbers of complaints despite high volume practices and high patient or procedural risk factors may identify best practices that can be disseminated for the benefit of others.

This study has several limitations. We did not assess the link between complaints and actual malpractice experience in this department. However, the significant association between patient complaints and malpractice claims has been well established elsewhere.9,11,12,15 A second limitation is that our postoperative complaint phone call system and database are limited to ambulatory patients. Additionally, not all complaints could be traced to a named anesthesiologist. For example, patients may report meeting several anesthesia providers and may only report a complaint regarding “one of them” without being able to recall the individual’s name. Also, because of the close relationship between surgery, perioperative nursing, and anesthesiology, complaints about nonanesthesia team members may have contaminated the anesthesiology data. Although we focused on attending anesthesiologists, some patients may not differentiate between anesthesiologists, certified registered nurse anesthetists, and other providers associated with their procedures. Complaints counted in this study unambiguously identified the anesthesiologist as the person associated with the concern. For the pediatric patient population, we were unable to ascertain and explore the gender of the survey respondent, while in the adult population, there were few nonpatient respondents, making that factor likely irrelevant. Last, the population studied consisted only of ambulatory patients who typically have fewer comorbidities and undergo less complex surgeries compared with the inpatient population. These limitations, however, should have been evenly distributed among the entire physician and patient populations studied and, therefore, should not have produced variation in the differences in patient complaint frequency or type.

In conclusion, patient complaints about the provision of anesthesia care are not evenly distributed among anesthesiologists, and certain patient and procedural factors contribute to increased complaint risk. For pediatric patients, older patient age and case delay were each associated with increased complaint risk whereas general anesthesia was associated with decreased risk. For adult patients, greater provider years in practice and calls made to the patient as opposed to a family member increased complaint risk whereas male gender and general anesthesia were associated with decreased risk. When these risk factors are included in regression models, evidence of heterogeneity of complaint rates among anesthesiologists is no longer apparent. However, complaints remain a significant predictor of malpractice risk, and physicians should recognize the importance of addressing complaints in a coordinated manner, for whatever reason they occur. Further investigation into how and why factors we have identified are associated with greater complaint risk may reveal potential interventions to decrease complaints, which in turn may decrease malpractice claims and improve patient and professional satisfaction.

Back to Top | Article Outline

DISCLOSURES

Name: J. Matthew Kynes, MD.

Contribution: This author helped design and conduct the study, analyze the data, and write the manuscript.

Attestation: J. Matthew Kynes has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Conflicts of Interest: The author has no conflicts of interest to declare.

Name: Jonathan S. Schildcrout, PhD.

Contribution: This author helped analyze the data and write the manuscript.

Attestation: Jonathan S. Schildcrout has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Conflicts of Interest: The author has no conflicts of interest to declare.

Name: Gerald B. Hickson, MD.

Contribution: This author helped design the study, analyze the data, and write the manuscript.

Attestation: Gerald B. Hickson has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Conflicts of Interest: The author has no conflicts of interest to declare.

Name: James W. Pichert, PhD.

Contribution: This author helped design and conduct the study, analyze the data, and write the manuscript.

Attestation: James W. Pichert has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Conflicts of Interest: The author has no conflicts of interest to declare.

Name: Xue Han, MS.

Contribution: This author helped analyze the data and write the manuscript.

Attestation: Xue Han has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Conflicts of Interest: The author has no conflicts of interest to declare.

Name: Jesse M. Ehrenfeld, MD, MPH.

Contribution: This author helped design and conduct the study.

Attestation: Jesse M. Ehrenfeld has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Conflicts of Interest: The author has no conflicts of interest to declare.

Name: Margaret W. Westlake, MLS.

Contribution: This author helped analyze the data and write the manuscript.

Attestation: Margaret W. Westlake has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Conflicts of Interest: The author has no conflicts of interest to declare.

Name: Tom Catron, PhD.

Contribution: This author helped design and conduct the study, analyze the data, and write the manuscript.

Attestation: Tom Catron has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Conflicts of Interest: The author has no conflicts of interest to declare.

Name: Paul St. Jacques, MD.

Contribution: This author helped design and conduct the study, analyze the data, and write the manuscript.

Attestation: Paul St. Jacques has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Conflicts of Interest: Paul St. Jacques has equity interest in Acuitec, LLC Acuitec, LLC markets a commercial version of the software which was used to collect the data for the study.

This manuscript was handled by: Franklin Dexter, MD, PhD.

a Consumer Complaint Reports and Statistics. Available at: http://www.bbb.org/us/statistics. Accessed February 7, 2012.
Cited Here...

Back to Top | Article Outline

REFERENCES

1. Department of Health and Human Services.CMS Interpretive Guidelines for Hospital Conditions of Participation Update. 2008 Washington, DC Department of Health and Human Services
2. Allen LW, Creer E, Leggitt M. Developing a patient complaint tracking system to improve performance. Jt Comm J Qual Improv. 2000;26:217–26
3. Hain PD, Pichert JW, Hickson GB, Bledsoe SH, Hamming D, Hathaway J, Nguyen C. Using risk management files to identify and address causative factors associated with adverse events in pediatrics. Ther Clin Risk Manag. 2007;3:625–31
4. Hsieh SY. The use of patient complaints to drive quality improvement: an exploratory study in Taiwan. Health Serv Manage Res. 2010;23:5–11
5. Jonsson PM, Øvretveit J. Patient claims and complaints data for improving patient safety. Int J Health Care Qual Assur. 2008;21:60–74
6. Moore IN, Pichert JW, Hickson GB, Federspiel C, Blackford JU. Rethinking peer review: detecting and addressing medical malpractice claims risk. Vand L Rev. 2006;59:1175–206
7. Parry J, Hewage U. Investigating complaints to improve practice and develop policy. Int J Health Care Qual Assur. 2009;22:663–9
8. Pichert JW, Hickson GB, Moore IN Using Patient Complaints to Promote Patient Safety: The Patient Advocacy Reporting System (PARS). 2008 Bethesda, MD Agency for Healthcare Research and Quality (AHRQ):421–30
9. Cydulka RK, Tamayo-Sarver J, Gage A, Bagnoli D. Association of patient satisfaction with complaints and risk management among emergency physicians. J Emerg Med. 2011;41:405–11
10. Hickson GB, Clayton EW, Entman PB, Sloan FA. Obstetricians’ prior malpractice experience and patients’ satisfaction with care. JAMA. 1992;267:1583–87
11. Hickson GB, Federspiel CF, Blackford J, Pichert JW, Gaska W, Merrigan MW, Miller CS. Patient complaints and malpractice risk in a regional healthcare center. South Med J. 2007;100:791–6
12. Hickson GB, Federspiel CF, Pichert JW, Miller CS, Gauld-Jaeger J, Bost P. Patient complaints and malpractice risk. JAMA. 2002;287:2951–7
13. Levinson W, Roter DL, Mullooly JP, Dull VT, Frankel RM. Physician-patient communication. The relationship with malpractice claims among primary care physicians and surgeons. JAMA. 1997;277:553–9
14. May M, Stengal B. Who sues their doctors? Law Soc Rev. 1990;24:105–20
15. Stelfox HT, Gandhi TK, Orav EJ, Gustafson ML. The relation of patient satisfaction with complaints against physicians and malpractice lawsuits. Am J Med. 2005;118:1126–33
16. Stimson CJ, Pichert JW, Moore IN, Dmochowski RR, Cornett MB, An AQ, Hickson GB. Medical malpractice claims risk in urology: an empirical analysis of patient complaint data. J Urol. 2010;183:1971–6
17. Vincent C, Young M, Phillips A. Why do people sue doctors? A study of patients and relatives taking legal action. Lancet. 1994;343:1609–13
18. Heidegger T, Husemann Y, Nuebling M, Morf D, Sieber T, Huth A, Germann R, Innerhofer P, Faserl A, Schubert C, Geibinger C, Flückiger K, Coi T, Kreienbühl G. Patient satisfaction with anaesthesia care: development of a psychometric questionnaire and benchmarking among six hospitals in Switzerland and Austria. Br J Anaesth. 2002;89:863–72
19. Myles PS, Williams DL, Hendrata M, Anderson H, Weeks AM. Patient satisfaction after anaesthesia and surgery: results of a prospective survey of 10,811 patients. Br J Anaesth. 2000;84:6–10
20. Caljouw MA, van Beuzekom M, Boer F. Patient’s satisfaction with perioperative care: development, validation, and application of a questionnaire. Br J Anaesth. 2008;100:637–44
21. Cass NM. Medicolegal claims against anaesthetists: a 20 year study. Anaesth Intensive Care. 2004;32:47–58
22. Kain ZN. The National Practitioner Data Bank and anesthesia malpractice payments. Anesth Analg. 2006;103:646–9
23. Chanthong P, Abrishami A, Wong J, Herrera F, Chung F. Systematic review of questionnaires measuring patient satisfaction in ambulatory anesthesia. Anesthesiology. 2009;110:1061–7
24. Fung D, Cohen MM. Measuring patient satisfaction with anesthesia care: a review of current methodology. Anesth Analg. 1998;87:1089–98
25. Herrera FJ, Wong J, Chung F. A systematic review of postoperative recovery outcomes measurements after ambulatory surgery. Anesth Analg. 2007;105:63–9
26. Chung F, Un V, Su J. Postoperative symptoms 24 hours after ambulatory anaesthesia. Can J Anaesth. 1996;43:1121–7
27. Kleinpell RM. Improving telephone follow-up after ambulatory surgery. J Perianesth Nurs. 1997;12:336–40
28. Tong D, Chung F, Wong D. Predictive factors in global and anesthesia satisfaction in ambulatory surgical patients. Anesthesiology. 1997;87:856–64
29. Le May S, Hardy JF, Taillefer MC, Dupuis G. Patient satisfaction with anesthesia services. Can J Anaesth. 2001;48:153–61
30. Mukherjee K, Pichert JW, Cornett MB, Yan G, Hickson GW, Diaz JJ Jr. All trauma surgeons are not created equal: asymmetric distribution of malpractice claims risk. J Trauma. 2010;69:549–54 discussion 554–6
31. Hickson GB, Pichert JW, Federspiel CF, Clayton EW. Development of an early identification and response model of malpractice prevention. Law Contemp Probl. 1997;60:7–29
32. Self SG, Liang KY. Asymptotic properties of maximum-likelihood estimators and likelihood ratio tests under nonstandard conditions. J Am Stat Assoc. 1987;82:605–10
33. Stram DO, Lee JW. Variance components testing in the longitudinal mixed effects model. Biometrics. 1994;50:1171–7
34. Carroll KN, Cooper WO, Blackford JU, Hickson GB. Characteristics of families that complain following pediatric emergency visits. Ambul Pediatr. 2005;5:326–31
35. Burroughs TE, Waterman BM, Cira JC, Desikan R, Claiborne Dunagan W. Patient satisfaction measurement strategies: a comparison of phone and mail methods. Jt Comm J Qual Improv. 2001;27:349–61
© 2013 International Anesthesia Research Society