Are Patient Self-Reports Adequate for Monitoring Health?

A major challenge in accurately diagnosing and treating patients is the quality of their self-reported data. Every patient’s experience with their health must be respected, but often the medical information and history they relay does not correlate with laboratory results and examinations. This definitely does not mean that patients are unengaged about the status of their health or lying. More likely, they are unaware of the progression of their illnesses, do not have the most up-to-date information about their health, or fall victim to their own biases and outsider perception about their conditions. Hence, the accuracy of self-reports depends on the patients’ knowledge of the relevant health information, the ability to recall it, and the willingness to report it to their providers.

One study measured how patients report on chronic illnesses such as hypertension. Comparisons of self-reported health information with physical examinations showed that patients largely underestimate the prevalence of hypertension. The accuracy of these reports also correlated with factors like age, education, when the respondents’ most recent health exam was, and cognitive abilities.

Similarly, researchers in China found that in lesser developed communities, many people are unaware of their conditions, and from a public health standpoint, these self-reports are leading to an underestimation of the prevalence of hypertension and diabetes. In more developed communities or provinces, self-reported data is considered to be more reliable with biomedical data.

With so many compounding variables, how can we make sure patients are fully informed and equipped to evaluate their health? Moreover, how can providers balance analyzing the accuracy of patient reports while also trusting and validating the patient’s experience?

Let’s take the example of medication intake. A typical conversation at an annual check-up visit may go like this:

Doctor: Are you still taking these medications? And how regularly are you able to?

Patient: Of course, I take the recommended dose daily.

In reality, it is completely possible that the patient may be embarrassed or hesitant to admit to their provider that they often miss a few pills. They may also have forgotten the many times they did not take their medication. Usually, this lack of continuity steadily increases over time in patients who have chronic conditions such as dyslipidemia, hypertension, or diabetes. Since medication adherence is absolutely essential in optimizing health outcomes and regulating treatment in chronic illness, misinformation can lead to a misrepresentation of disease management.

Research shows that self-report medication adherence accuracy varies with how physicians phrase their questions, how much patients have to recall, and how long the recall period is. Factors that also complicate recall for patients and lead to inaccuracies include the characteristics of the recalled phenomenon (recency, attributes, complexity) and context of the recalled phenomenon (salience, patient experience, mood).

As a result of these elements, self-reports overestimate adherence behavior. Most evidence indicates that self-report adherence measures show moderate correspondence to other adherence measures and can significantly predict clinical outcomes. The quality of self-report adherence measures may be enhanced through efforts such as:

• The use of population-specific adherence measures with validation data
• Reducing social desirability bias with open language and statements that normalize nonadherence
• Employing computer services to measure adherence measures rather than face-to-face data collection to alleviate social desirability concerns and improve quality

However, these initiatives are not foolproof in preventing skewed self-reports, especially when it comes to more stigmatized behaviors that have serious health consequences.

For example, 1 in 5 Americans admit lying to their doctor about how much alcohol they consume. Another study found that there is a significant level of opioid use underreporting in hospitalized patients and healthy individuals. When it comes to substance abuse, these inaccuracies can lead to worsening addictions and health outcomes. Medical professionals may be able to detect these omissions of truth through certain indicators such as elevated enzyme levels or high blood pressure; but having access to a patient’s full history and lifestyle is important to make connections about their symptoms and conditions.

Similarly, for smoking cessation, in a study analyzing the self-reports of cardiac patients, 25% inaccurately reported having quit smoking. This underreporting was more prevalent among patients who received face-to-face counseling and had an “intermediate” education level. Without corrections to these differences and biases, many people will continue to be more susceptible to disease without even realizing the damage they are doing.

The quality of self-reports can clearly be influenced by bias and stigma, and while we can try to correct the methods used to collect these reports, nothing beats raw biomedical data. Results from laboratory tests and physical exams are more reliable than patient reports, but it is not very feasible for this data to be collected even weekly or monthly, let alone daily.

Technologies like Somatix’s SafeBeing™ are providing a solution. Through passively monitoring an individual’s behavior, these smartbands collect powerful clinical insights needed to evaluate a patient’s conditions: sleep (quantity and quality), medication intake, smoking, drinking, activity levels, and more. Using these tools, there are proven reduction rates for those in addiction recovery, improvements in cessation efficacy, and lower readmission rates.

Human nature should not hinder patients from receiving the best possible care. Harnessing the power of technology can negate the risks of errors and disparities in self-reporting.