Safety of care
Identifying the risks in medical practice
Originally published May 2015
The growing emphasis on healthcare safety has spawned a dramatic increase in scientific knowledge about improvements in patient care and managing risk in clinical situations. Identifying the nature and sources of risk are important elements in quality and safety improvements. For its part, the CMPA contributes to a more thorough understanding of the nature of medico-legal risk in medical care through its analyses of closed cases.
The CMPA’s medico-legal cases help to identify potentially high-risk areas of practice. Notably, in the last 5 years surgery was the area of practice with the highest risk of being involved in a medico-legal matter. Almost half of closed CMPA cases in this period related to a surgical procedure. Diagnostic error, which is often difficult to identify and isolate from other associated issues, comprised 16% of the cases; medication was primarily involved in 8% of cases; and another 8% involved obstetrics.
Risk can also be identified by observing which groups of physicians are most often involved in legal actions. Physicians’ level of medico-legal risk is influenced primarily by the types of patients and conditions they typically treat, as well as by the procedures they perform.
The severity of patient outcomes can also determine the level of risk. Labour and delivery, and procedures on the brain or spinal column are high-risk because of the potential for catastrophic patient outcomes when complications arise. For this reason, specialists performing these procedures including obstetrician-gynaecologists, neurosurgeons, and orthopaedic surgeons are considered high-risk.
Beyond medico-legal case data
While medico-legal case data can provide a useful snapshot of potential risk, evidence-based patient safety research and advanced analytics delve deeper into the complex reasons for patient safety incidents — at both the system and individual healthcare provider levels. Whether an organization is large or small, system refers to a healthcare organization’s work environment (including resources), processes of care, protocols, and fail-safe systems.
What is a fail-safe system?
The CMPA defines a fail-safe system as protocols, procedures, or systems in hospitals, offices, and clinics designed to prevent or mitigate errors. For example a process to follow up on the results of lab tests and diagnostic imaging studies so abnormal findings are flagged and reliably acted on is a fail-safe system. Another example is the use of checklists in surgery.
It is widely recognized that most patient safety incidents result from flawed systems and failures in processes. Because these incidents are often complex, involving multiple players and contributing factors, systems thinking approaches, such as patient safety incident analysis (formerly called root cause analysis), can be useful for identifying the potential multiple contributory causes of unintended outcomes. Indeed, the Institute of Medicine states that the most effective way to reduce error in medicine is to focus on system-level improvements.
While reducing system-level risks may yield the greatest improvements in patient safety, healthcare providers are personally responsible for patient safety in the direct care they provide to their patients. In some cases, a healthcare provider’s personal characteristics place them at higher risk of a legal action or regulatory authority (College) complaint. In addition to attributes such as interpersonal skills, communication style, and technical skills and knowledge, clinical decisions are also shaped by human factors that include cognitive and affective biases.
These biases are a major area of study in the field of risk management and patient safety. Cognitive biases (i.e. distortions and short-cuts in thinking) and affective biases (i.e. intrusion of the physician’s feelings) may interfere significantly with reaching a correct diagnosis. Common cognitive biases associated with error include, anchoring (fixating on one particular symptom, sign, or piece of information, or a particular diagnosis); premature closure (accepting a first plausible initial diagnosis before making a reasonably complete verification); and attribution error (a form of stereotyping that involves explaining a patient’s condition based on their disposition or character). Because these biases are not always easy to spot, physicians must be vigilant to their existence and potential impact in individual cases.
As seen in the following case example, patient safety incidents often involve a combination of system and human factors.
Case example: Inadequate assessment highlights risk stemming from cognitive bias
A middle-aged woman presents to the emergency department (ED) with severe left-sided chest pain that intermittently radiates to her back, left arm, and abdomen, and is worse on inspiration. A resident assesses her and notes possible guarding of the abdomen. He orders medication for pain as well as investigations that include troponin and D-dimer tests, an electrocardiogram, and chest and abdominal X-rays. Her history includes recent personal stressors, constipation, and successful treatment for primary bone cancer 10 years prior.
An hour later, the on-call ED physician assesses the patient and reviews the notes of the resident and nurses, and the test results. The abdominal X-ray suggests moderate fecal loading and other test results are unremarkable but for a slightly elevated D-dimer. The physician concludes that her symptoms are related to her constipation. No longer in pain, the patient is discharged with a prescription for a laxative and advised to return to the ED if she has ongoing concerns.
Six months later, the patient is diagnosed with a cancerous tumour in her lung. She complains to the physician’s regulatory authority (College) when she discovers that the report of the chest X-ray taken at the time of her previous ED visit had identified a small pleural effusion in the same area of the lung where her tumour was ultimately found.
The College expert finds that the ED physician’s assessment of the patient was inadequate. While an elevated D-dimer level can be non-specific, in light of the patient’s presenting complaints and history of cancer (however remote), further tests should have been conducted to rule out potentially serious conditions (e.g. pulmonary embolism). The expert suggests that the ED physician’s focus on an abdominal cause for the patient’s symptoms likely led him to discount other potential red flags. In addition, the expert states that there should be a system in place for reviewing test results flagged as abnormal. This system should include reconciling the test results with the patient’s medical record before making decisions on follow-up.
The ED physician is required to meet with the College to discuss implementing changes to his practice to avoid similar missed diagnoses.
Medical care analytics at the CMPA
The CMPA uses the following analytic practices to collect and understand medico-legal case data with the goal of contributing to safe medical care:
- Clinical coding: Case details, including contributing factors, are re-constructed with codes that comply with international standards. This facilitates comparison with other organizations.
- Graphing: Diagrams show relationships between events and contributing factors. This identifies patterns of risk.
- Mapping: “Process-of-care” frameworks pinpoint where specific problems occur (e.g. in surgery looking at the pre-operative, operative, and post-operative phases of care). This identifies opportunities for individual and system improvements.
- Modelling: Statistical models seek to quantify the influence of certain factors on outcomes. This helps in understanding the contributors to patient harm.
The bottom line
Being aware of high-risk areas as identified through research, and appreciating the diverse influences of individual and system factors on patient safety can help physicians recognize the potential for unique risks in their practice environment. Additionally, physicians may find the following sources of information beneficial when attempting to identify and mitigate risk areas in their practice:
- patient safety incident and near miss reporting systems
- quality improvement reviews
- patient complaints
- morbidity and mortality rounds
- performance or comparative data from the Canadian Institute for Health Information audit results
- Accreditation Canada assessments