International Journal of Surgery 35 (2016) 44e50
Contents lists available at ScienceDirect
International Journal of Surgery journal homepage: www.journal-surgery.net
Original Research
Clinical audit system as a quality improvement tool in the management of breast cancer Chellappa Vijayakumar a, Nanda Kishore Maroju a, *, Krishnamachari Srinivasan a, K. Satyanarayana Reddy b a b
Department of Surgery, Jawaharlal Institute of Postgraduate Medical Education and Research (JIPMER), Pondicherry, 605006, India Department of Radiotherapy, Jawaharlal Institute of Postgraduate Medical Education and Research (JIPMER), Pondicherry, 605006, India
h i g h l i g h t s In-hospital delay can occur at several points in the clinical care pathway. Audit-based feedback is a reliable method of improving timeliness in the care of breast cancer patients. Resource deficits should be recognized and addressed in process improvement.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 20 May 2016 Received in revised form 14 August 2016 Accepted 11 September 2016 Available online 12 September 2016
Introduction: Quality improvement is recognized as a major factor that can transform healthcare management. This study is a clinical audit that aims at analysing treatment time as a quality indicator and explores the role of setting a target treatment time on reducing treatment delays. Materials and methods: All newly diagnosed patients with breast cancer between September 2011 and August 2013 were included in the study. Clinical care pathway for breast cancer patients was standardized and the timeliness of care at each step of the pathway was calculated. Data collection was spread over three phases, baseline, audit cycle I, and audit cycle II. Each cycle was preceded by a quality improvement intervention, and followed by analysis. Results: A total of 334 patients with breast cancer were included in the audit. The overall time from first visit to initiation of treatment was 66.3 days during the baseline period. This improved to 40.4 and 28.5 days at the end of Audit cycle I and II, respectively. The idealized target time of 28 days for initiating treatment was achieved in 5, 23.5, and 65.2% of patients in the baseline period, Audit cycle I, and Audit Cycle II, respectively. There was improvement noted across all steps of the clinical care pathway. Conclusion: This study confirms that audit is a powerful tool in quality improvement programs and helps achieve timely care. Gains achieved through an audit process may not be sustainable unless underlying patient factors and resource deficits are addressed. © 2016 IJS Publishing Group Ltd. Published by Elsevier Ltd. All rights reserved.
Keywords: Quality improvement Audit Breast cancer Target time Tracking
1. Introduction Quality assurance is a newer field of study in healthcare management that aims to improve patient outcomes, patient experience and treatment cost. Surgical audit is an important tool in ensuring quality care. It is particularly important in providing continuous monitoring and feedback in the implementation of clinical care pathways. Timeliness of care is an important indicator
of quality, and is found to be deficient even in health care systems among the developed nations [1e3]. Breast cancer is a common malignancy and has a fairly well established treatment pathway, and is an ideal case for quality improvement studies [4]. A key factor reflecting the quality of care in the management of patients with breast cancer is the timeliness of care [5,6]. This study is a clinical audit that aims at analyzing treatment time as a quality indicator and attempts to explore the role of setting a target treatment time on reducing delays in treatment [7].
* Corresponding author. E-mail address:
[email protected] (N.K. Maroju). http://dx.doi.org/10.1016/j.ijsu.2016.09.011 1743-9191/© 2016 IJS Publishing Group Ltd. Published by Elsevier Ltd. All rights reserved.
C. Vijayakumar et al. / International Journal of Surgery 35 (2016) 44e50
2. Materials & methods This was a quality improvement audit conducted from September 2011 to August 2013 in the breast cancer unit of a tertiary referral hospital. All newly registered patients during this study period with suspected breast cancer were enrolled in the study. Auditing of data was done in patients who were confirmed to have breast cancer on histopathology. The study was approved by the Institutional Research Monitoring and Ethical Committees. All patients gave informed consent for inclusion in the study. The hospital had a standardized clinical care pathway for breast cancer management (Fig. 1). 2.1. Phases of the audit The audit involved three phases, baseline phase, audit cycle I, and audit cycle II. 2.1.1. Baseline phase (pre-tracking) The baseline phase included data collection from September 2011 to August 2012. This phase also involved evaluation of the reasons for delay in the treatment of breast cancer at our center. 2.1.2. Audit meeting I (brainstorming) At the end of the baseline phase, a brainstorming session was held, which was attended by faculty and residents from the departments of surgery, pathology, radio diagnosis and imaging, radiation oncology and medical oncology. The timeliness of care across the clinical pathway was analyzed and possible solutions to reduce delays were discussed. 2.1.3. Audit cycle I This cycle of data collection was conducted over a 6-month period from October 2012 to March 2013. The brainstorming session and the interventions agreed upon were the only difference in the conditions of the baseline and audit cycle I. 2.1.4. Audit cycle II This was the final cycle of data collection conducted over a 5month period from April 2013 to August 2013. This phase incorporated setting of artificial targets for initiation of treatment. In this phase we divided the overall target time into smaller blocks where individual actions could be achieved. The overall target to start treatment was fixed at 28 days to give allowance of a week for each of the four main steps in the treatment pathway. Weekly reminders were given to all surgical units as well as radiation oncology and medical oncology. The basis for giving these reminders was to ensure that all members involved in the pathway were aware of all patients and their respective position in the clinical care pathway. It also enabled the coordinator to identify patients who were lagging behind on the clinical care pathway.
45
Treatment time was recorded during each of the audit cycles. The treatment time was recorded for the following segments: (1) first hospital visit to histological diagnosis; (2) diagnosis to cancer conference; (3) cancer conference to initiation of treatment (Surgery/chemotherapy/radiotherapy). An artificial target time of 28 days was set as the time provided from first out-patient visit to initiation of treatment. This time was further divided and allotted for each activity in the care pathway. 2.2. Statistical analyses Descriptive statistics were used to compare timeliness across the audit cycles. Fisher's exact test and one-way ANOVA were used to compare the differences of values between the audit cycles. A two-sided P-value < 0.05 was considered statistically significant. 3. Results A total of 397 patients with breast lumps suspicious of malignancy were enrolled in the study. All 397 cases were followed up in different phases of the audit cycles. Thirty-seven of these cases were found to be benign, 15 patients were lost to follow up, and 17 patients expired during the study period (Fig. 2). Majority of the patients diagnosed with carcinoma breast were above 45 years of age (62.3%). About 69.2% patients were in postmenopausal age group, 37.2% patients were from a low economic status (annual income < INR 24,000 per month), and 64.9% patients had not completed secondary schooling. Half of the patients (48.8%) presented to the out-patient department (OPD) 3 months after they noticed the lump. The most common reason for delayed presentation was the lump being painless (60.8%) (Table 1). 3.1. Treatment times 3.1.1. First visit to confirmation of diagnosis The time taken to receive a final histological confirmation of diagnosis was 23.1 days in the baseline period. This time was reduced to 15 days in audit cycle I, and 10.4 days in audit cycle 2. Analysis revealed that the principal delays occurred in the timing of the core needle biopsy, the histological reporting and assessment of hormonal assessment. These delays were addressed during each of the audit cycles to reduce the time to achieve diagnosis. 3.1.2. Confirmation of diagnosis to multi-department conference The time taken for the patient to be discussed in the multidepartment conference was 40.5, 25.4, and 18.1 days, in the baseline, audit cycle I, and audit cycle II, respectively. Analysis revealed that the principal reasons for delay in this period were owing to delayed appointments for imaging and bone scanning for staging purposes. The interventions identified were to encourage department level prioritizing of investigations, as well as the willingness
Fig. 1. Clinical care pathway.
Fig. 2. Study flow diagram.
C. Vijayakumar et al. / International Journal of Surgery 35 (2016) 44e50
47
Table 1 Demographic factors. Parameters
Baseline (N ¼ 100)
Audit cycle I (N ¼ 112)
Audit cycle II (N ¼ 122)
Age > 45 years Post-menopausal Income < INR 2000/month Education < High School >3 month history Reasons for delayed presentation Painless Small size Suspected to be lipoma
67 59 39 79 45
79 83 44 73 53
82 89 42 65 62
49 26 25
72 23 17
82 30 10
to discuss cases in the cancer conference before bone scans (the investigation with maximum waiting time) were available. 3.1.3. Multi-department conference to initiation of treatment Breast cancer patients were either considered for primary surgery or neoadjuvant chemotherapy. A small proportion of patients were considered for primary radiotherapy. The overall time taken to initiate treatment was 2.8 days in the baseline phase, 2.2 days in audit cycle I, and 1.8 days in audit cycle II. On considering these times with respect to the different treatment options, there was a significant improvement in the time taken to initiate chemotherapy. However, patients scheduled for primary surgery had to wait longer in Audit cycle II (Table 2). 3.2. Target time Target times were defined to identify the exact number of patients who experienced delays in management. The target time for treatment initiation from the first visit was fixed at 28 days, with durations specified for each of the steps in the treatment pathway. Only 5% of the patients in the baseline phase achieved this target time. This improved significantly to 23.5% in Audit cycle I, and 65.2% in Audit cycle II. The target time of 14 days from the first visit to histological confirmation was achieved in 97.5% of the patients in Audit cycle II compared to 62.5% in audit cycle I and 16% at baseline. Target time of 7 days between histological confirmation to final treatment plan (7 days) was achieved in 72% of patients in baseline, 73.2% in Audit cycle I, and 90.9% in audit cycle ll. The target time of 7 days between treatment plan to starting of treatment was achieved in 19% in the baseline group, 42.9% in audit cycle I, and 61.5% in audit cycle ll. All improvements were statistically significant (Table 3). 3.3. Reasons for failure in achieving target time Incomplete histological report was the most common reason for failure in achieving target time in the baseline period (52.6%), and the second most common reason in audit cycle l (22.4%). The
frequency of incomplete histology was significantly reduced in audit cycle II. The delay due to imaging was the most common reason for delay in audit cycle l. Failure rate in achieving primary surgery (BCS and MRM) within the target period increased in audit cycle II (47.6%) compared with audit cycle I (14.1%) and base line (10.5%). Delay in primary surgery was the most common reason for failure in audit cycle ll. Delay in starting chemotherapy within 28 days was the second most common reason for failure in audit cycle II (16.7%) (Table 4).
4. Discussion This study attempted to analyze breast cancer treatment as an example of a process, which could be improved. Of the several components of interest, the one from the patient's point of view is effective care, curative or palliative. At the next level are indices that reflect patient satisfaction. Timeliness of care is increasingly being referred to as a key quality indicator. Studies have also demonstrated that timeliness significantly impacts treatment outcomes as well. And above all in any quality initiative, timeliness is the easiest parameter to measure. Herein, we analyzed the impact of audit cycles on timeliness of care in the management of breast cancer [8]. All cases at our hospital were managed in the Comprehensive Breast Cancer Center, that, though not governed by the EUSOMA guidelines, fulfils the prescribed requirements [9]. Breast cancer patients are initially worked up in the surgical outpatient department before they are routed to the breast center. However, quality issues arise owing to the demographics and logistics of a heavily populated developing country. These include patient populations with suboptimal awareness, and healthcare systems that are typically overburdened. The period of baseline data collection was important in terms of analyzing the care pathway against the existing system and people related conditions. This was essential for any process improvement intervention to succeed. The baseline data, the collection of which was spread over almost a year, demonstrated an inconsistent and prolonged duration between the patient's first visit to the hospital
Table 2 Treatment time (in days). Treatment period (Days)
Baseline N ¼ 100 Mean ± SD
Audit cycle I N ¼ 112 Mean ± SD
Audit cycle II N ¼ 122 Mean ± SD
First visit to Core-needle biopsy Core-needle biopsy to complete HPE report First visit to complete HPE report HPE Report to Multidepartment conference First visit to Multidepartment conference Treatment plan to treatment initiation First visit to treatment started
9.3 ± 8.9 13.7 ± 7.9 23.1 ± 12.1 6.7 ± 8.2 32.3 ± 17.8 2.8 ± 5.9 66.3 ± 24.3
6.6 ± 7.9* 8.4 ± 3.9* 15 ± 8.2* 5.8 ± 5.9 20.8 ± 10.1* 2.2 ± 4.8 40.4 ± 18.5*
5.8 ± 1.5 8.1 ± 1.7 10.4 ± 2.2** 4.3 ± 4.1 14.7 ± 4.5** 1.8 ± 1.8 28.5 ± 10.1**
*P < 0.05, compared to Baseline; **P < 0.05, compared to Audit cycle l. HPE, histopathological examination.
48
C. Vijayakumar et al. / International Journal of Surgery 35 (2016) 44e50
Table 3 Target time. Treatment period
First visit to core needle biopsy Core needle biopsy to HPE report First visit to HPE report HPE report to treatment plan Treatment plan to treatment initiation First visit to treatment initiation
Designated target time (days)
7 7 14 7 7 28
Number of patients who achieved target time (%) Baseline (n ¼ 100)
Audit cycle I (n ¼ 112)
Audit cycle II (n ¼ 122)
54 (54) 10 (10) 16 (16) 72 (72) 19 (19) 5 (5)
81 54 70 82 48 27
121 (99.5)** 36 (29.5)** 119 (97.5)** 111 (90.9)** 75 (61.5)** 80 (65.2)**
(72.3)* (48.2)* (62.5)* (73.2) (42.9)* (23.5)*
*P < 0.05, compared to baseline; **P < 0.05, audit cycle II compared to audit cycle I. HPE, histopathological examination.
Table 4 Reasons for failure in achieving target time.
No. of patients that failed to meet target time Repeat core needle biopsy Incomplete HPE Report Metastatic work up Delayed Surgery Fitness for anesthesia Delayed Chemotherapy Delayed Radiotherapy
Baseline (n ¼ 100)
Audit cycle I (n ¼ 112)
Audit cycle II (n ¼ 122)
95 14 (14.7%) 50 (52.6%) 13 (13.7%) 10 (10.5%) 3 (3.2%) 4 (4.2%) 1 (1.1%)
85 10 (11.7%) 19 (22.4%)* 32 (37.7%)* 12 (14.1%) 0 (0%) 3 (3.5%) 9 (10.6%)*
42 3 (7.1%) 2 (4.8%)** 4 (9.5%)** 20 (47.6%)** 0 (0%) 7 (16.7%)** 6 (14.3%)
Percentages calculated with respect to the number of patients that failed to meet target time. *P < 0.05, compared to baseline; **p < 0.05, compared to audit cycle I. HPE, histopathological examination.
and the initiation of treatment. This period was broken down into specific segments to analyze the cause of such delay, as well as to understand the strengths and weaknesses of the hospital system. 4.1. Resource deficit The baseline data demonstrated that the hospital services were overburdened with a heavy patient and work load. The Surgery Outpatient Department (OPD), which was managed by two consultants and ten residents, was visited by an average of 200 patients every day. The department of Pathology screens 20,000 tissue samples for histopathological analysis every year with a consultant strength of five. Since every report is counter checked by a consultant, there is an unavoidable delay in reporting of biopsies. Availability of Hormone Receptor kits is a problem in our set up as they are expensive and have a limited shelf life. The department of Radiotherapy admits an average of 2500 patients every year for radiation therapy. Its status as a Regional Cancer Center implied that the two linear accelerators available in the development catered to a population of 10 million, amounting to 20 times the ratio seen in developed countries. The department of Radio diagnosis images about 6000 patients every year for CT Scanning alone. The department has an acute shortage of consultants, residents and technicians and despite their best efforts, the timeliness in terms of quicker turnover of patients, did not show any improvement, demonstrating clearly the effect of resource deficit. In all these situations there is clearly a factor of system overload. It is well known that systems that are deficient in human resources, infrastructure or finances are hamstrung in terms of delivering quality care [10]. 4.2. Patient related factors The importance of patient related factors cannot be overlooked in analyzing delays in treatment. The delay is usually seen either in presenting to the hospital or in ensuring compliance with the
clinical pathway. While the quality of counseling determines compliance to an extent, the ability of patients to understand and adhere to an agreed schedule is also important. In our study population, 90% of patients were of low socioeconomic status, with one third of them being below the poverty line. Two thirds of patients did not complete secondary schooling. Significantly, more than half of all patients presented 3 months after they were aware of a lump in their breasts. Socioeconomic factors and personal factors play a strong role in contributing to the delay in treatment [10,11]. 4.3. Baseline data and brainstorming session The process of brainstorming process threw up interesting answers and insights. The results of the brainstorming session were summarized and circulated among the treating teams. All the teams involved in treating breast cancer patients were included in this session. It was felt that the process of discussing the problem and finding solutions was an important first step in initiating process change. The solutions listed in Table 1 were only the ones that were clearly articulated, and other minor issues may have been solved without directly addressing them. 4.4. Assessment at the end of audit cycle I Data collected during Audit cycle I that followed the brainstorming session demonstrated significant improvement in the time taken at each of the steps. The only step where the improvement was not statistically significant was the delay between formulating a treatment plan and starting actual treatment. However, this time was not prolonged even in the baseline period. The highest quantum of improvement was noted in the time taken to report core needle biopsies. Reports were now available in eight to nine days in most cases. But the ‘range’ still showed cases being reported after three weeks. The total time lapse between the first visit till the initiation of treatment that was at an average of 66 days per patient, improved to 40 days indicating that the brainstorming
C. Vijayakumar et al. / International Journal of Surgery 35 (2016) 44e50
session had contributed to a greater than 30% reduction in treatment delay. 4.5. Audit cycle II The second audit cycle was an example of a closely monitored treatment process. In established breast care centers it is the task of the breast cancer care coordinator to ensure that unnecessary delays are avoided. In our setup, there were multiple surgical units that cared for patients with breast cancer, in addition to managing other general surgical conditions, including acute surgery. The likelihood of a patient dropping off the watch of a unit was therefore quite high. The department did not have a dedicated breast cancer coordinator. The resident in charge of the audit assumed the role of the coordinator and printed out, as well as emailed weekly reminders that listed patients at different stages of work up against the target time for each patient. All patients likely to miss their target times were highlighted. This was an intensive effort and was continued for 20 weeks. The data collected at the end of Audit cycle II demonstrated a significant improvement in treatment time with the overall time to start treatment dropping to 28 days as opposed to the baseline delay of 66 days and the delay of 40 days after completion of Audit Cycle I. We relooked at all timelines through the treatment target times set by us to measure the ability to meet these targets through the different stages of the audit cycles and found a predictable increase in the number of patients meeting target times as the intensity of tracking increased. In the baseline data, only 5% of patients met the overall treatment target of 28 days. This improved to 23% at the end of Audit cycle I, and to 80% at the end of Audit Cycle II. The increase in the number of patients not meeting target times in audit cycle II for primary surgery or chemotherapy was explained by an increase in patient burden in that cycle. Since the first two steps in the care pathway were significantly faster, the number of patients who were ready for treatment increased. However, there was no increase in the operating time, or hospital beds, thus leading to greater waiting time for primary treatment. 4.6. Hawthorne effect and incentivizing
49
4.7. Limitations The gains achieved in terms of improving timeliness of care with brainstorming and weekly reminders were very encouraging. Such an improvement is unlikely to be sustainable unless other visible and invisible barriers to timely care are improved [11]. Primary among these are patient awareness, resource matching and team building activities, which were not addressed by the current study. This study also failed to focus on other quality indicators. 5. Conclusion This study confirms that audit is a powerful tool in quality improvement programs and assists in achieving timeliness of care in the management of patients with breast cancer. Weekly reminders were necessary to consolidate the gains achieved through brainstorming. Gains achieved through an audit process may not be sustainable unless patient factors and resource deficits are addressed. Ethical approval This study was approved by the institutional review and ethical board. Sources of funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Author contribution Chellappa Vijayakumar: Data acquisition, drafting the manuscript. Nanda Kishore Maroju: Study design, data analyses, drafting and critical revision. Srinivasan Krishnamachari: Study design, data analysis and critical revision. K S Reddy: Data analysis and critical revision. All authors have approved the final version of the manuscript. Conflicts of interest
The improvement from Audit cycle I to Audit cycle II demonstrates that setting target times and intense tracking constitute an important component of any Quality Improvement Process. While brainstorming and awareness does improve performance, the resulting improvement is limited. In comparison a system of constant feedback and reminders is much more effective in making a system work. This is explained by the Hawthorne effect, meaning performance of a unit or person improves with the knowledge that the unit or person is being evaluated. In addition, comparison with other patients on the reminder sheet also serves the purpose of providing competition. There is an amount of incentivizing in setting up target times and rewarding achievers and naming defaulters. This appeared to work across all departments and across all care providers. Similar experience has been noted in several other settings [8]. A number of factors influence delivery of timely care to patients with breast cancer. Resource management and patient awareness are important determinants in the quality of care provided. There is also a clear overload of existing resources in the hospital, both in terms of manpower and infrastructure. This audit aimed at correcting delay outside the ambit of these two limitations, and focused mainly on achieving goals in terms of target times within the clinical care pathway.
None of the authors have any conflicts of interest to report. Guarantor Nanda Kishore Maroju. References [1] P.A. Van Dam, G. Verheyden, A. Sugihara, et al., A dynamic clinical pathway for the treatment of patients with early breast cancer is a tool for better cancer care: implementation and prospective analysis between 2002-2010, World J. Surg. Oncol. 11 (2013) 70, http://dx.doi.org/10.1186/1477-7819-11-70. [2] A. Ponti, M.P. Mano, M. Tomatis, et al., Audit system on Quality of breast cancer diagnosis and Treatment (QT): results of quality indicators on screendetected lesions in Italy, 2011-2012, Epidemiol. Prev. 39 (2015) 40e47. PMID: 26405775. [3] E.A. McGlynn, S.M. Asch, J. Adams, et al., The quality of health care delivered to adults in the United States, N. Engl. J. Med. 348 (2003) 2635e2645. PMID: 12826639. [4] H. Bao, F. Yang, S. Su, et al., Evaluating the effect of clinical care pathways on quality of cancer care: analysis of breast, colon and rectal cancer pathways, J. Cancer Res. Clin. Oncol. 142 (2016) 1079e1089, http://dx.doi.org/10.1007/ s00432-015-2106-z. [5] J. van Hoeve, L. de Munck, R. Otter, J. de Vries, S. Siesling, Quality improvement by implementing an integrated oncological care pathway for breast cancer patients, Breast 23 (2014) 364e370, http://dx.doi.org/10.1016/
50
C. Vijayakumar et al. / International Journal of Surgery 35 (2016) 44e50
j.breast.2014.01.008. [6] H.M. Schachter, V. Mamaladze, G. Lewin, et al., Many quality measurements, but few quality measures assessing the quality of breast cancer care in women: a systematic review, BMC Cancer 6 (2006) 291. PMID: 17176480. €nen, Y. Collan, Diag[7] E. Ermiah, F. Abdalla, A. Buhmeida, E. Larbesh, S. Pyrho nosis delay in Libyan female breast cancer, BMC Res. Notes 5 (2012) 452, http://dx.doi.org/10.1186/1756-0500-5-452. [8] K. Sharma, A. Costas, L.N. Shulman, J.G. Meara, A systematic review of barriers to breast cancer care in developing countries resulting in delayed patient presentation, J. Oncol. 2012 (2012) 121873, http://dx.doi.org/10.1155/2012/
121873. [9] A.R.M. Wilson, L. Marotti, S. Bianchi, et al., The requirements of a specialist Breast centre, Eur. J. Cancer 49 (2013) 3579e3587, http://dx.doi.org/10.1016/ j.ejca.2013.07.017. [10] T.G. Hislop, S.R. Harris, J. Jackson, et al., Satisfaction and anxiety for women during investigation of an abnormal screening mammogram, Breast Cancer Res. Treat. 76 (2002) 245e254. PMID:12462385. [11] J. Ferlay, H.-R. Shin, F. Bray, D. Forman, C. Mathers, D.M. Parkin, Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008, Int. J. Cancer 127 (2010) 2893e2917, http://dx.doi.org/10.1002/ijc.25516.