By David Zitner, July 30, 2020
The COVID-19 pandemic has brought renewed attention to the challenges facing Canadian health care. Widespread community testing for the disease will prove crucial in the fight against COVID. Yet, in the absence of testing capacity, we will need to better identify the common problems in communities and better assess which interventions are most helpful and which ones are most harmful. In other words, we need to better connect our health care activities to changes in health status for our patients.
However, one particular challenge is our inability to capture, in real time, information about the health and sickness of Canadians. Health systems in Canada do not routinely record, assess, and report the benefits and harms of care. Consequently, it is difficult for health services administrators and policy-makers to learn which health care activities are most helpful, harmful, or merely a waste of time. COVID-19 has certainly magnified the need to better understand the efficacy of our health care activities, but our shortcomings in such assessments have long been a problem in our health system.
In 1994, the Federal, Provincial and Territorial Deputy Ministers of Health unanimously and enthusiastically endorsed a report they commissioned titled “When Less is Better,” especially its recommendations to routinely report on access (waiting times) for care and the outcomes of care. Canadian health organizations spend billions, yet they cannot tell you how many people were healthier or sicker following care, nor how many people are waiting, how long they wait, and the consequences of extended waits for care. It seems they just do not know how to do it.
However, there are pockets of excellence. For example, cardiology groups across the country can tell you the consequences of waiting times related to the attributes of patients. They can stratify patient risk and allocate the timing of services based on the patient health status. Those at risk of harm while waiting are treated more quickly than those less likely to suffer from delayed diagnosis or treatment.
Usually patients who receive treatment know if they were better or worse, helped or harmed, from the tests or treatment. At every visit each clinician assesses a patient’s health in order to know whether to begin treatment, continue treatment or discontinue existing medication.
It would be simple to routinely collect health status information from each encounter and combine the information from multiple patients and encounters. That could be used to track the health of individual patients, combine the information, and learn overall which interventions are most likely to be helpful or harmful and who is most likely to be harmed by waits for treatment.
Such information would also give us a better sense of the impact that pandemic preparedness might have on the health of Canadians who might be waiting for treatment. After all, many hospitals prepared for a possible COVID-19 surge that fortunately did not occur, resulting in excess, unused, clinical capacity that could have been used by non-COVID patients. Clearly, some people must have been harmed because they could not receive, or did not seek, timely and appropriate care. In the absence of health status information, we simply do not have the data to make such an assessment.
We can and must do better. Sometimes patients visit doctors to learn what to do to maintain their health. Usually, the main reasons people see doctors is because they want to feel better, do more, or live longer. In other words, the dimensions of health are comfort, function and likelihood of dying. Each of these is measurable.
Comfort: Estimating Subjective Pain and Feelings
Individuals usually have an overall impression of how they feel. They also know if they have aches or pains, or discomfort in one or another part of their body. Comfort measures are necessarily subjective. Patients report how they feel. Feelings are not observable by the doctor, although demeanor is.
Doctors are usually unable to objectively confirm what a person feels and usually must accept what the patient describes as an accurate reflection of what the patient is experiencing.
The elements that contribute to measures of comfort include:
- Measures of overall feelings of well-being.
- Organ or site-specific measures – for example, headaches or a painful knee.
- Psychological well-being – is an individual satisfied with their life? Is the person happy and content with their relationships? Is the person happy and content with their financial and social state?
Effective treatment improves overall comfort. Medicine has not succeeded when a drug controls pain from an ankle sprain but produces moderate or severe abdominal pain and nausea or even worse bleeding from the stomach. It is important to measure and record both.
Measures of comfort indicate what individuals perceive about themselves. An external observer has no way of knowing whether the person who reports they are in pain is telling the truth or lying. However, it is in everyone’s best interests if health providers assume that people, looking for health advice, are telling the truth.
Most of us do not need formal scales to help us understand how we feel. However, health researchers and clinicians often want to know whether particular treatments or changes in lifestyle were worthwhile or not, and they use formal measures to assess changes in comfort associated with care.
Assessments of overall feelings of well-being or comfort have questions most of us would expect. For example, “Do you feel healthy enough to do the things you would like to do?” Some use scales and ask you to rate how you feel on a numerical scale from one to 10. Others ask about your sleep and diet habits, or particular questions about various organs or systems. For example, do you feel stomach sick? Do you have a good appetite? Modern information techniques make it easy to record, store and use this kind of information.
Clinicians use a variety of scales to record overall function as well the function of particular organs and joints. This is recorded, one way or another, in patients’ charts.
Function measures are necessarily objective. Measures of function indicate what an individual can do. An external observer can easily tell whether you can run a mile in about seven minutes, the range of motion of your arms or legs, your visual acuity, and how well you can remember a list of names or numbers.
Everyone, usually, has an idea of their overall function and the function of particular body parts, and these personal assessments, usually, but not always, relate closely to objective measures.
Measures of function, include overall measures represented by questions like:
- Can you perform the normal activities of daily living?
- Are you able to care for yourself or do you need help?
- Can you walk, and how quickly, up or down a hill?
- Can you feed yourself?
Or specific measures related to one or another system, such as:
- What is the range of motion of your knee? Of your elbow?
- When the doctor taps your knee, is there a normal knee reflex (which provides a measure of neurological function)?
- Can you remember a list of names or numbers? How long is the list of numbers you can remember?
This kind of information is also recorded after every encounter.
Estimates of comfort and function along with laboratory investigations all contribute to estimates of life expectancy.
Clinicians and lay people regularly estimate the health of colleagues, neighbors and even passersby we meet on the street. Sometimes we say, “He looked to be at death’s door,” meaning he looked extremely sick and likely to die. These informal assessments are often correct, but sometimes inaccurate.
Doctors order laboratory investigations, blood tests, and images including X-ray and magnetic resonance imaging (MRI) to further refine their estimates of health. The results of tests contribute to estimates of a person’s overall health, including mortality. People with blood test evidence of poor organ function are at higher risk of death compared with people with evidence of normal or healthy organ function. Patients with pneumonia, gasping for breath and with an abnormal heart rate, are at higher risk of dying compared with patients with pneumonia whose heart rate is close to normal and who are not gasping for breath.
Objective clinical measures, laboratory investigations and medical images produce objective information from many sources, including an analysis of bodily fluids (blood, mucous, saliva, sweat, and urine), feces, and medical images including x-rays, and nuclear medicine scans.
A variety of algorithms are useful and with reasonable accuracy predict what patients can expect from care and life expectancy. One classic example is the APACHE (Acute Physiology and Chronic Health Evaluation) score, which collects information about patients in intensive care units and produces an estimate of the probable outcomes in critically ill patients.
All of this information is also recorded in every patients’ chart and can be used to develop predictive models.
Health systems that routinely collect information about the health of patients should also use this information in real time (in the same way that Amazon, Costco, Facebook, and Google do) to learn about the health of patients in communities and to link health care activities to the results of care.
Although organizations capture information about the admission and discharge diagnoses of patients, they do not routinely include information about changes in health, comfort, function, and life expectancy associated with care. Cardiac care and rehabilitation medicine groups are the exceptions that show that measuring health and recording health outcomes can be done.
Cardiac care and rehabilitation medicine routinely capture, record, and use information about health and changes in health associated with care, to allocate resources and to identify opportunities to improve.
Timely information about health status is also useful to identify when many individuals in a community are afflicted by common symptoms to identify new outbreaks of disease.
For COVID-19, routine community testing is important. In the absence of testing capacity, it is important to identify the most common problems in communities and the interventions that are most helpful or harmful. The same methods – linking health care activities to changes in health status – also enable clinicians and administrators to learn which health care activities help, which harm, and which are just a waste of time and money.
Dr. David Zitner, a retired family physician, was a Professor and the founding director of the Graduate Program in Health Informatics at Dalhousie University.