Key performance indicators for monitoring the integrated care pathway in breast cancer: the E.Pic.A. project

Introduction

In recent decades, the Italian National Health System (INHS) has undergone a continuous and rapid process of innovation that has led to the achievement of remarkable clinical goals; as one example, the increase in life expectancy in Italy has now given it a top position in world rankings, and mortality rates for major diseases have also been reduced (1). This process of innovation, in large part due to the introduction of new drugs on the market, has also led to an increase in public drug expenditure, especially in the field of oncology. Indeed, in 2015, at €2,372 million, oncology drugs represented the highest pharmaceutical costs (10.5%) by the INHS (2).

In the face of mounting restrictions in healthcare expenditure that predictably will not undergo significant increases in the coming years, it seems reasonable to expect that no new financial resources will be made available to the INHS for expenditures on anti-cancer drugs, while the number of new diagnoses of cancer continues to grow (3, 4). It thus becomes increasingly important to place greater attention on increasing the efficiency of how healthcare resources are allocated along Integrated Care Pathways (ICPs), with a view on eliminating waste and saving costs. According to a recent economic analysis, waste of resources (e.g. overuse of inappropriate health interventions) increases healthcare expenditure by more than 20% (5).

To guarantee the survival of universalistic health system in such a picture where growing health needs are accompanied by rising costs and risk of wastages, the key point is to give value to healthcare that, according to Sir M. Gray, involves optimal allocation of resources (allocative value), together with a radical recovery of efficiency in a context that guarantees an unchanged or, if possible, greater effectiveness (technical–professional value) in full compliance with the individual patient’s value system (personal value) (6).

In Italy, in consideration of its impact in terms of incidence (28%) and prevalence (23% of all cancer patients) in women (7), breast cancer (BC) is a paradigmatic oncological condition where one can study value along the care pathway.

In addition, BC is a complex clinical condition whose management requires a multidisciplinary approach. This has led to centralise the ICP by specialised BC units (BUs). BUs allow patients to face BC with the certainty that they will be followed by a multidisciplinary team of dedicated specialists, treated according to the highest standards and accompanied throughout the disease journey. The BU is therefore an excellent model of intervention, thanks to which adherence to shared treatment standards has improved, with outcomes of the overall treatment process that are standardised and multidimensional (8).

With the aim of increased efficiency, it becomes essential to measure the performance of the various activities needed for the ICP in BC. In the past, several multidisciplinary groups of professionals in the BU have defined a minimum set of quality indicators to measure the clinical efficacy of the ICP of BC (9, 10).

In the present study, we report the methodology developed by the E.Pic.A. (Economic Appropriateness of an Integrated Care Pathway) multidisciplinary panel (composed of clinicians, health management physicians, methodologists, IT experts), which is aimed at constructing an analytic process that allows evaluation of clinical and economic inappropriateness in BC by defining key performance indicators (KPIs) (11).

Methods

In accordance with current guidelines of the Associazione Italiana di Oncologia Medica (AIOM) (12) and the National Comprehensive Cancer Network (13), the E.Pic.A. multidisciplinary panel was asked to identify and define a series of KPIs that can be utilised to evaluate the potential inappropriateness of healthcare services in the ICP in BC. In order for the evaluation process to be homogeneously applied by all hospitals, the KPIs identified by the E.Pic.A. multidisciplinary panel had to be easily measurable using currently available and commonly used databases (e.g. administrative databases, hospital discharge card, etc.).

In particular, each of the KPIs identified had to be “reliable”, i.e. subject to the lowest number of systematic errors possible, and “representative”, i.e. characterised by a close relationship with the phenomenon that it intends to evaluate (14). The KPIs also had to meet two other requirements: “accessibility”, or be easily measurable using the available information sources, and “operativity”, or be easily applicable in the context to be evaluated, allowing identification of critical issues and monitor interventions (14). For each KPI defined, the E.Pic.A. multidisciplinary panel was asked to: (i) provide a definition, (ii) explain the methodology behind its calculation, and (iii) indicate the target for evaluation.

The present study describes only analytic methodology and does not contain any medical information about an identifiable patient. The findings will be shared with other hospitals and will be disseminated to the scientific community at conferences.

Key performance indicators

A total of seven KPIs were identified to measure inappropriateness of healthcare services provided in the ICP in BC: three for examinations, two for surgery and two for treatment. The three KPIs related to examinations were aimed at estimating the inappropriateness of diagnostic tests performed in the pre- and post-index breast surgery phases. The two KPIs for surgery measured the inappropriateness of re-surgery during the time elapsed from the index breast surgery. The two KPIs for treatment measured inappropriateness when referred to the time of administration of adjuvant therapy (chemotherapy and/or hormonal therapy) and radiotherapy after the index breast surgery.

Since the date of execution of the index breast surgery (and relative index hospitalisation) represents the time reference for calculation of each of the seven KPIs, it was essential to provide a definition that also included identification criteria. First, since it is possible to identify new BC patients undergoing breast surgery each year, the E.Pic.A. multidisciplinary panel established that the time period during which breast surgery would be identified is the calendar year (January 1 to December 31). The index breast surgery procedure was then defined as the first hospital admission for (total or partial) resection surgery in the year of observation characterised by the concomitant presence of one of the diagnostic­ codes according to the International Classification of Diseases 9^th revision Clinical Modification (ICD-9-CM) and one of the surgical codes listed in Tables I and II. Thus, from the index breast surgery identified, the following were excluded:

− all male patients;

− patients less than 18 years of age at the index breast surgery;

− in prevalent cases, at least one hospitalisation for BC (ICD-9-CM V10.3 or 174* or 233.0) as the main or secondary diagnosis in the period between 3 years and 6 months prior to the date of the index breast surgery;

− in cases with another neoplasm, at least one hospital admission (ICD-9-CM 140*-173* or 176*-195 or 200-208* or V.10*excluding V10.3) for a main or secondary diagnosis different from BC in the 3 years preceding and 6 months after the index breast surgery;

− presence of metastatic disease at hospitalisation (ICD9-CM 197*, 198*, 199*) for the index breast surgery or during hospitalisation in the three years preceding the index hospital admission.

Lastly, to determine staging of the BC, which is fundamental in the measurement of KPIs, the E.Pic.A. multidisciplinary panel adopted the tumour, node, metastases (TNM) coding described in the AIOM 2017 guidelines (15) listed in Table III; preference was given to pathological TNM staging, while clinical staging was used as an alternative. For each of the seven KPIs, the definitions, reference targets and calculation methods used by the E.Pic.A. multidisciplinary panel are provided.

Discussion and conclusions

The methodology developed by the E.Pic.A. multidisciplinary panel in the present study evaluated the ICP of BC patients, and identified a set of KPIs that examine some of the most critical factors related to the efficiency of a process or particular activity (11).

It is likely that one of the biggest problems with the INHS is the cascading effect of the decision-making process combined with poor overall accountability of the system. The public payor is different from those who prescribe (clinicians) and, without adequate planning and structuring of the INHS, the process of allocating resources often becomes complex and flawed, making it impossible to efficiently use the available resources. Therefore, providing tools to guide the treatment pathway for both economic (general managers) and clinical (clinicians) decision makers represents the attempt to bridge the gap between those who prescribe and those who are in charge of economic resources. In this light, clinical/administrative data are useful for constructing an accurate representation of real-word practice, in which all the different stakeholders who provide healthcare services can make their own evaluations in a systematic manner.

As already indicated herein, within a context characterised by ever-increasing healthcare needs, it would be impracticable to believe that increasing resources will become available in the near future. Hence, the E.Pic.A. multidisciplinary panel developed a methodology to identify inappropriate (or potentially inappropriate) activities that allow, within a specified ICP, reduction of waste so that new resources can be made available. The application of this methodology should allow achievement of short-term (reduction of inappropriateness), medium-term (ensure clinical outcomes while reducing waste) and long-term (reduce waiting lists and increase availability of healthcare technologies) objectives. This would thus increase the value of the healthcare services provided by the INHS. In fact, by improving allocation of resources in relation to burden of illness and need for healthcare, it is possible to increase the performance of both hospitals and professionals (technical–professional value), in line with the highest expectations of each individual patient (personal value).

Compared to the core set of parameters defined by EUSOMA (European Society of Breast Cancer Specialists) (9), composed of 17 indicators to evaluate quality of treatment pathways for BC, the seven KPIs defined by the E.Pic.A. multidisciplinary panel are mainly aimed at identification of potential inappropriateness related to providing unnecessary services during the ICP of BC. Specifically, the methodology attempts to identify potential areas with low “value” in terms of utilisation of resources, and reducing the associated costs. An initial and partial application of this methodology has recently been carried out at the Istituto Scientifico Romagnolo per lo Studio e la Cura dei Tumori IRST of Meldola (Forlì, Italy) (17). The study estimated that the hospital provided 2,516 inappropriate examinations (from January 2010 to June 2016) with a total cost of about €580,000.

The present methodology will be applied at regional (Agenzia Regionale di Sanità Toscana, Istituto per lo studio, la prevenzione e la rete oncologica Toscana, and Azienda Sanitaria Locale CN2 Piemonte) and local levels (Istituto Scientifico Romagnolo per lo Studio e la Cura dei Tumori IRST of Meldola, Forlì, Italy and Istituto di Sanità Pubblica Sezione di Igiene dell’Università Cattolica del Sacro Cuore, Rome, Italy) with the aim to provide a more complete picture of inappropriateness and waste, allowing resources to be redirected towards more valuable interventions/examinations for patients with BC (not only), and close the gap between correct use of resources and achieving the best clinical results.

Disclosures

Financial support: This study was made possible by an educational grant from Roche S.p.A.

Conflict of interest: The authors did not receive financial support for the study. AP is a full-time employee of Roche Diagnostics S.p.A.