Brian K. Clinton, MD, PhD

Should patients be allowed to read their charts? A patient may reason, “It is my medical record, I paid for it, and I have the right to see it,” or the more vexing, “My insurance company read my chart, so I should have the right to read it!” Yes, the patient likely signed a form acknowledging that insurance companies have special access under the Health Insurance Portability and Accountability Act of 1996 (HIPAA), but at that time the patient had little awareness of a future desire to see his or her own chart.

A psychiatrist may feel it is counter-therapeutic to disclose the raw material in the chart. Perhaps it will inspire shame, hopelessness, or anger. A psychiatrist may be insecure about revealing poor record-keeping habits or, more subtly, may feel discomfort with the notion that reading the chart allows the patient to glimpse into the psychiatrist’s mind.

There has been substantial controversy about whether patients should be allowed to read their psychiatric record. Traditionally, patients have not had the legal right, but this has changed in recent decades, and federal law now strongly supports a patient’s right to view the chart on request. Here I review the ethical and legal factors involved in this challenging situation.

Ethical arguments about a patient’s right to see the psychiatric chart balance autonomy (the patient’s decision whether to view the chart) and respect for persons (the patient’s right to have maximal autonomy) against the principles of non-maleficence and beneficence (both as judged by the clinician). Also relevant is the duty of veracity/truth-telling that may be compromised not only by lying but also by omission (eg, not disclosing the chart).

Does releasing the chart to a patient risk causing harm to the patient or someone else? If so, what level of risk may be acceptable? Could reading the chart be helpful to the patient? These are difficult questions to answer when a carefully created therapeutic alliance hangs in the balance between psychiatrist and patient. This alliance is the soil for productive psychological work, and psychiatrists are loath to imperil it by turning over a copy of the psychiatric record.

It should be recognized, however, that a patient’s request to read the chart is a point of double jeopardy for the treatment relationship, which is endangered whether access is granted or denied. At this point, we must remind ourselves that it is the patient, and not the therapeutic alliance, that is to be prioritized. The long-term benefits of strengthening patient autonomy should not be hindered without good reason.

How do we predict the risks and benefits of granting or denying access to the chart? Several related studies offer guidance.

 Examination of the content of psychiatric case notes revealed that up to 80% contained elements that were potentially puzzling, offensive, alarming, or upsetting to patients.

 Studies of the impact of patients’ review of their charts have been conducted primarily in acute inpatient settings, usually during active treatment and with direct supervision. A substantial percentage (35% to 50%) of inpatients felt more pessimistic or upset after reading their records, but there was also a consensus that little or no substantial harm was involved.

Most inpatients who read their records reported that they felt better-informed and more involved in treatment.

 Some patients with psychotic diagnoses thought it demystified the medical record.

 Staff members reported that the studies changed their charting practice (eg, by minimizing upsetting notations and psychotic diagnoses), and they observed that important communications sometimes failed to be charted.

Outpatient data are scarce; however, one study reported benign results when carefully screened outpatients read their records in a clinically supportive setting.

 These studies do not take into account the subpopulations of patients who spontaneously make requests to see their chart, although one retrospective study suggests a preponderance of character disorders and contentious treatment relationships.

 The available literature is limited in broad applicability, but it suggests that while clinicians fear the risks of chart disclosure, the practice in supervised settings has been benign or even beneficial.

The HIPAA Privacy Rule strongly supports a patient’s right to inspect and obtain a copy of the medical record. However, there are exceptions to this legal right that are relevant to mental health care. Psychiatrists need to know when the legal right exists, since it may take precedence over one’s ethical stance on the issue. Of note, when state law and federal law are in conflict, the law that prevails is generally whichever more stringently protects a patient’s privacy or more liberally grants a patient access to the chart. According to HIPAA, a patient may obtain a copy of his chart via 3 routes with each requiring successively more effort: (1) a treating provider grants access; (2) the treating provider denies access, but an uninvolved “reviewer” or review committee grants access; or (3) a court overrules any denials of access.

On what grounds can clinicians deny chart access? HIPAA states that a clinician may deny access on “reviewable” grounds if the disclosure creates a likely risk. “Reviewable” indicates that a patient may always request an appeal when denial is based on assessments of risk. Interestingly, this risk threshold is set at 2 levels, depending on who is at risk. Chart access may be denied to a mentally competent patient if disclosure is “reasonably likely to endanger the life or physical safety” of the individual (or another person). Thus, the risk threshold for denying chart access to a competent patient is set high and is actually intended to prevent injury, suicide, physical violence, and homicide by the patient.

Clinicians should note that a concern of emotionally bruising the patient is an inadequate legal rationale to deny access to the chart. In cases that may be likely to cause some emotional distress, federal law favors patient autonomy over the physician’s instinct “to do no harm.”

If the at-risk party is another person (not the patient), denial may be acceptable at a lower risk threshold defined as “reasonably likely to cause substantial harm to such other person.” This lower threshold for denial also applies to a mentally incompetent patient when a personal representative requests the chart (for example, a parent of a minor child, or a guardian of an incompetent adult). The term “substantial harm” is an inclusive criterion that may refer to physical, emotional, or psychological harm. Regardless of the applicable threshold of harm, an attempt to precisely understand the imprecise language “reasonably likely” may lead a clinician down a legal rabbit hole filled with case law. It is worthwhile to consult a trusted attorney and/or colleague when a clinician finds it challenging to decide whether any anticipated violence is “reasonably likely” or somewhat less likely.

Other exclusions to a patient’s right to view his chart are considered “non-reviewable” and are not guided by a threshold of harm. HIPAA specifies that a treating clinician may deny a patient access if the records sought meet the legal definition of “psychotherapy notes”; the chart is a forensic evaluation; the chart is part of a consented, ongoing research study; or the chart compromises the confidentiality of another party (eg, reference to another patient or to a confidential nonclinician informant, such as a parent who provides sensitive personal history).

HIPAA appears to allow physicians a secondary chart (ie, the psychotherapy note) to be used for more sensitive information. However, psychotherapy notes are not completely inaccessible, since they may become evidence in a legal proceeding. Nonetheless, HIPAA allows 2 distinct sets of notes-the medical progress notes versus the psychotherapy notes. HIPAA narrowly defines psychotherapy notes as a set of notes kept in a separate location from the medical record, which include personal notes and observations and do not include the information that is appropriately held in the medical progress notes.

This distinction is important because, as legally defined, a psychotherapy note must omit much of what a clinician might include, and it is not as comprehensive as might be imagined. The psychotherapy note is a useful clinical tool to jog memory on more personal information, references to other people, and clinical speculation. This sensitive information is better kept out of the patient’s medical record, both because it compromises patient privacy with insurance companies and because it may create an unnecessary obstacle to releasing the medical record to the patient.

The official patient chart notes should include objective findings such as progression of symptoms, diagnosis, treatment plans, and services rendered.

 This document may include the minimum necessary amount of documentation required for ensuring quality clinical care and obtaining insurance reimbursement. It is also judicious to include some description of the patient’s positive personal qualities and strengths.

A records request in a private practice setting has the advantage that the clinician has had a large amount of control over what has been placed in the chart and knows the patient well. Ideally, the documentation is tasteful and will be minimally surprising to the patient. The request may feel emotionally charged because it may be occurring in the setting of some discord within a lengthy, perhaps ongoing treatment relationship. An ongoing therapeutic relationship will assist with risk assessment and mitigation of these risks. If a request for records is made, it is recommended that the process of denial or disclosure is included as part of the psychotherapeutic work.

Institutional records (eg, inpatient or emergency department charts) present unique challenges to disclosure. In some institutional settings, review committees and institutional protocols may dictate documentation guidelines and the disclosure decision process. However, many requests for institutional records come directly to the involved clinician. It might be assumed that chart release to a patient after discharge is relatively simple because there is no therapeutic alliance to preserve. This is not necessarily the case.

The patient chart is often a conglomeration of notes produced by an attending psychiatrist, covering psychiatrists, consultants, trainees, social workers, and nurses; it may even contain documents obtained from the patient (eg, a suicide note). Documentation may be excessive to convey the case clearly to covering clinicians and insurance companies. A medical trainee may include psychodynamic speculations or may begin progress notes with unforgiving clinical language, such as “oddly related 29-year-old unemployed, unmarried, childless female with a severe personality disorder arriving after a failed suicide attempt.” The chart may include extensive collateral information from family and outpatient clinicians, some of which may have been offered confidentially and undisclosed to the patient in crisis. For these reasons, the institutional chart frequently meets criteria for denial. It involves a high-risk patient and content that may be surprising and disturbing, and it likely holds information that compromises the confidentiality of others.

Moreover, there may be no information about the patient’s current mental state. Risk assessment is thus confounded, and there is little ability to mitigate the impact of chart disclosure. If an inpatient chart is problematic, I recommend an initial denial or an offer to provide a summary of the hospital stay (a maneuver allowed by HIPAA). Another alternative is to offer release directly to another clinician or an attorney who agrees to meet with the patient.

In the course of usual practice, it is likely that a patient will request to see the chart. Given that the current state of the law supports patient access, it is my opinion that clinicians should attempt to optimize documentation so that there are as few barriers as possible to safely releasing psychiatric medical records upon a patient’s request. The Table provides a suggested strategy to guide clinical practice when a patient asks to see his medical chart.

Dr Clinton is Associate Professor of Psychiatry at Columbia University Medical Center and Associate Director of Inpatient Psychiatry Service on 9 Garden North at New York-Presbyterian Hospital/Columbia University Medical Center in New York. He reports no conflicts of interest concerning the subject matter of this article.

 Ross SE, Lin C. The effects of promoting patient access to medical records: a review. 

 Schwartz HI, Rachlin S. Patient access to mental health records: impact on clinical practice. 

 Sergeant H. Should psychiatric patients be granted access to their hospital records? 

 Bloch S, Riddell CE, Sleep TJ. Can patients safely read their psychiatric records? Implications of freedom of information legislation. 

 Crichton P, Douzenis A, Leggatt C, Hughes T. Are psychiatric case-notes offensive? 

 Stein EJ, Furedy RL, Simonton MJ, Neuffer CH. Patient access to medical records on a psychiatric inpatient unit. 

 Miller RD, Morrow B, Kaye M, Maier GJ. Patient access to medical records in a forensic center: a controlled study. 

 Bernadt M, Gunning L, Quenstedt M. Patients’ access to their own psychiatric records. 

 Parrott J, Strathdee G, Brown P. Patient access to psychiatric records: the patients’ view. 

 Simonton MJ, Neuffler CH, Stein EJ, Furedy RL. The open medical record: an educational tool. 

 Essex B, Doig R, Renshaw J. Pilot study of records of shared care for people with mental illnesses. 

 Bernstein RA, Andrews EM. Response strategies for chart requests from psychiatric inpatients. 

 Appelbaum PS, Gutheil TG. Clinical Handbook of Psychiatry and the Law. 4th ed. New York: Lippincott 

Articles

Federal law now strongly supports a patient’s right to view his or her psychiatric record on request. Here: a look at the ethical and legal issues.

Brian K. Clinton, MD, PhD

Articles

Ethical Issues in Disclosing to Patients: Should Patients Be Allowed to Read Their Charts?