Certainly a key component of FTR is delayed escalation of care. Johnston and colleagues identified key steps in the postoperative care escalation process in an academic medical center and potential reasons for failure at each step [9]. In the nursing level, some of the steps where escalation could potentially fail included the following: the patient failing to inform the nursing assistant of feeling unwell, failure of the nurse to attend to the patient or notice deterioration, failure to measure or document vital signs correctly, and failure to inform the junior resident about the deterioration. Clinical understaffing was determined to be the principal root cause of all these failures. At the junior resident level, potential points of care escalation failure included failure to perform an adequate history and physical, failure to initiate treatment, and failure to inform a senior resident of the deterioration. Root causes of these failures included hierarchical barriers to communicating to the senior resident, senior resident unavailability, and poor chart design or legibility. Lastly, at the senior resident level, failure to arrange definitive management was recognized as a potential point of escalation failure, with root causes of hierarchical barriers to communicating with the attending surgeon and lack of hospital resources. Proposals to address all the above root causes included increasing nurse-to-patient ratios, adding more permanent nursing staff, investment in electronic vital sign recording and medical records, removal of hierarchical barriers, increased use of smartphone technology, educating junior residents about the importance of prompt care escalation, development of a clear escalation protocol, development of guidelines defining appropriate levels of care according to diagnosis and physiological parameters, and ensuring adequate access to resources, such as operating rooms and surgical intensive care unit beds.

Further illustrating the importance of hospital systems and communication in the quality of postoperative care, Symons and colleagues analyzed postoperative care of 50 patients undergoing elective major general surgery and found that process failures accounted for 57 % of all preventable adverse events, and more than half of those failures were due to communication errors and delays in patient care [10].

To determine the relative importance of FTR as a reason why some hospitals have higher surgical mortality rates, Hyder and colleagues compared hospitals on the basis of mortality rates using data from the Nationwide Inpatient Sample [11]. They established five subpopulations as potential targets for mortality improvement: patients with one of the six AHQR FTR-qualifying complications, patients with high estimated preoperative risk, emergency surgery patients, elderly patients, and patients with diabetes. Comparing hospitals with high mortality rates to those with low mortality rates, mortality risk differences were greatest in the high-risk and FTR subpopulations . They found in simulations that optimizing outcomes in the FTR population could potentially reduce the mortality gap between the highest and lowest mortality hospitals by as much as 75 %.

Also making the connection between FTR and mortality, Wakeam and colleagues used rates of secondary complications after common index complications (a failure to arrest progression of complications) as a surrogate for FTR [12]. Using the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP ) data, they found a wide variation in the rate of complications among hospitals. Hospitals in the highest quintile of secondary complications after pneumonia, myocardial infarction, and surgical site infection had significantly greater mortality rates.

Also using ACS-NSQIP data, Ferraris and colleagues stratified patients according to their propensity for developing serious postoperative complications based on more than 50 demographic and preoperative clinical variables [13]. They then defined FTR as death after a serious postoperative complication and found that 88 % of patients with FTR were in the highest risk quintile, and 95 % were in the two highest risk quintiles. A single postoperative complication often preceded the development of multiple complications. As the number of complications increased, so did the FTR rate, as shown i n Fig. 4.1 [13]. More than two-thirds of the FTR patients had multiple postoperative complications, and the mortality rate increased exponentially with the number of complications. Interestingly, patients with surgical residents involved in their care had lower FTR rates.

In a recent review article, Johnston and colleagues found that the reported incidence of FTR in surgery patients varied between 8.0 and 16.9 % [14]. Several studies found that an increased incidence of FTR was associated with high mortality rates and low hospital volume. Illustrating the importance of patient factors in FTR, lower FTR rates were associated with patient age <70 years, absence of malignancy, and white ethnicity. The nature of the initial complication also affects FTR. Greater FTR rates were found in patients with medical (as opposed to surgical) complications, surgical site infections, DVT, pneumonia, sepsis, and gastrointestinal complications. Just as reported in other studies, hospital system issues affect the FTR rate. Multiple studies showed increased mortality rates when escalation of care was delayed. The reasons for such a delay included incorrect diagnosis, clinical inexperience, poor communication due to hierarchical barriers, fear of criticism, high workload, overconfidence, frequent interruptions, and clinician unavailability.

Using FTR as a safety and quality metric is not without flaw. FTR rates are calculated from administrative data sets and, thus, rely on accurate coding. Certain diagnoses, such as trauma and immunocompromised status, are often excluded from analysis [3]. Education, patient selection, and communication are only a few variables that result in failing to rescue a patient from a complication. Therefore, it is often very difficult to determine the exact cause of differences in FTR rates. FTR should be used as one of many quality metrics when assessing a program, hospital, or system. It is clear that FTR is an important indicator to provoke and assess change, but the exact impact on patient safety in the US healthcare system will be difficult to determine. While not currently part of the selected PSIs used by CMS to impact reimbursement, FTR in the coming years may be linked to financial consequences.

The Institute of Medicine reports To Err is Human: Building a Safer Health System and Crossing the Quality Chasm: A New Health System for the twenty-first Century increased attention to identifying and preventing medical errors and resulted in the implementation of many quality measures. Assessing the effectiveness of these interventions remains difficult. Ultimately, reporting overall mortality for surgical procedures is challenging since most reports are from single institutions, small groups of hospitals, or a single state. Gawande and colleagues looked at 30-day outcomes reported in the Nationwide Inpatient Sample and found that inpatient deaths were lower in 2006 than 1996 [15]. Similar improvements have been reported in the National Hospital Discharge Survey database , the National Surgical Quality Improvement Project , and other national health data registries [16–18]. The observed decrease in mortality, however, does not correlate directly with a decrease in adverse or potentially preventable events. Since the implementation and tracking of PSIs , there has been an overall decrease in FTR rates [19]. In contrast, the rates of some important PSIs have increased, namely postoperative thrombotic events, sepsis, respiratory failure, hospital-acquired infections, decubitus ulcer, and accidental puncture or laceration [15, 19].

While most PSI events are increasing, postoperative mortality rates are decreasing. The reason for this discordance is unknown, but several other observations offer plausible explanations. The decrease in FTR rates over this interval suggests improvement in education and communication , thus leading to earlier recognition of complications before progression to mortality [18]. Patients are living longer and have increasing comorbidities, and operating on higher risk patients leads to more postoperative complications [20]. Moreover, with increased attention to patient safety, more resources have been devoted to proper coding, leading to more reported complications.