Diagnosis of borderline personality disorder is based on a clinical assessment by a mental health professional. The best method is to present the criteria of the disorder to a person and to ask them if they feel that these characteristics accurately describe them. Actively involving people with BPD in determining their diagnosis can help them become more willing to accept it. Although some clinicians prefer not to tell people with BPD what their diagnosis is, either from concern about the stigma attached to this condition or because BPD used to be considered untreatable, it is usually helpful for the person with BPD to know their diagnosis. This helps them know that others have had similar experiences and can point them toward effective treatments.

In general, the psychological evaluation includes asking the patient about the beginning and severity of symptoms, as well as other questions about how symptoms impact the patient's quality of life. Issues of particular note are suicidal ideations, experiences with self-harm, and thoughts about harming others. Diagnosis is based both on the person's report of their symptoms and on the clinician's own observations. Additional tests for BPD can include a physical exam and laboratory tests to rule out other possible triggers for symptoms, such as thyroid conditions or substance abuse. The ICD-10 manual refers to the disorder as "emotionally unstable personality disorder" and has similar diagnostic criteria. In the DSM-5, the name of the disorder remains the same as in the previous editions.

Psychiatrists seek to provide a medical diagnosis of individuals by an assessment of symptoms, signs and impairment associated with particular types of mental disorder. Other mental health professionals, such as clinical psychologists, may or may not apply the same diagnostic categories to their clinical formulation of a client's difficulties and circumstances. The majority of mental health problems are, at least initially, assessed and treated by family physicians (in the UK general practitioners) during consultations, who may refer a patient on for more specialist diagnosis in acute or chronic cases.

Routine diagnostic practice in mental health services typically involves an interview known as a mental status examination, where evaluations are made of appearance and behavior, self-reported symptoms, mental health history, and current life circumstances. The views of other professionals, relatives or other third parties may be taken into account. A physical examination to check for ill health or the effects of medications or other drugs may be conducted. Psychological testing is sometimes used via paper-and-pen or computerized questionnaires, which may include algorithms based on ticking off standardized diagnostic criteria, and in rare specialist cases neuroimaging tests may be requested, but such methods are more commonly found in research studies than routine clinical practice.

Time and budgetary constraints often limit practicing psychiatrists from conducting more thorough diagnostic evaluations. It has been found that most clinicians evaluate patients using an unstructured, open-ended approach, with limited training in evidence-based assessment methods, and that inaccurate diagnosis may be common in routine practice. In addition, comorbidity is very common in psychiatric diagnosis, where the same person meets the criteria for more than one disorder. On the other hand, a person may have several different difficulties only some of which meet the criteria for being diagnosed. There may be specific problems with accurate diagnosis in developing countries.

More structured approaches are being increasingly used to measure levels of mental illness.

- HoNOS is the most widely used measure in English mental health services, being used by at least 61 trusts. In HoNOS a score of 0–4 is given for each of 12 factors, based on functional living capacity. Research has been supportive of HoNOS, although some questions have been asked about whether it provides adequate coverage of the range and complexity of mental illness problems, and whether the fact that often only 3 of the 12 scales vary over time gives enough subtlety to accurately measure outcomes of treatment.

The person's appearance, behavior, and history, along with a psychological evaluation, are usually sufficient to establish a diagnosis. There is no test to confirm this diagnosis. Because the criteria are subjective, some people may be wrongly diagnosed.

The previous edition of the Diagnostic and Statistical Manual of Mental Disorders, DSM IV-TR, defines histrionic personality disorder (in Cluster B) as:

The DSM-IV requires that a diagnosis for any specific personality disorder also satisfies a set of general personality disorder criteria.

Since the 1980s, Paula Caplan has been concerned about the subjectivity of psychiatric diagnosis, and people being arbitrarily “slapped with a psychiatric label.” Caplan says because psychiatric diagnosis is unregulated, doctors are not required to spend much time interviewing patients or to seek a second opinion. The Diagnostic and Statistical Manual of Mental Disorders can lead a psychiatrist to focus on narrow checklists of symptoms, with little consideration of what is actually causing the patient’s problems. So, according to Caplan, getting a psychiatric diagnosis and label often stands in the way of recovery.

In 2013, psychiatrist Allen Frances wrote a paper entitled "The New Crisis of Confidence in Psychiatric Diagnosis", which said that "psychiatric diagnosis… still relies exclusively on fallible subjective judgments rather than objective biological tests." Frances was also concerned about "unpredictable overdiagnosis." For many years, marginalized psychiatrists (such as Peter Breggin, Thomas Szasz) and outside critics (such as Stuart A. Kirk) have "been accusing psychiatry of engaging in the systematic medicalization of normality." More recently these concerns have come from insiders who have worked for and promoted the American Psychiatric Association (e.g., Robert Spitzer, Allen Frances). A 2002 editorial in the "British Medical Journal" warned of inappropriate medicalization leading to disease mongering, where the boundaries of the definition of illnesses are expanded to include personal problems as medical problems or risks of diseases are emphasized to broaden the market for medications.

With treatment, the majority of people with BPD can find relief from distressing symptoms and achieve remission, defined as a consistent relief from symptoms for at least two years. This longitudinal study tracking the symptoms of people with BPD found that 34.5% achieved remission within two years from the beginning of the study. Within four years, 49.4% had achieved remission, and within six years, 68.6% had achieved remission. By the end of the study, 73.5% of participants were found to be in remission. Moreover, of those who achieved recovery from symptoms, only 5.9% experienced recurrences. A later study found that ten years from baseline (during a hospitalization), 86% of patients had sustained a stable recovery from symptoms.

Patient personality can play an important role during the therapeutic process, leading to better clinical outcomes. Recent research has shown that BPD patients undergoing dialectical behavior therapy (DBT) exhibit better clinical outcomes correlated with higher levels of the trait of agreeableness in the patient, compared to patients either low in agreeableness or not being treated with DBT. This association was mediated through the strength of a working alliance between patient and therapist; that is, more agreeable patients developed stronger working alliances with their therapists, which in turn, led to better clinical outcomes.

In addition to recovering from distressing symptoms, people with BPD also achieve high levels of psychosocial functioning. A longitudinal study tracking the social and work abilities of participants with BPD found that six years after diagnosis, 56% of participants had good function in work and social environments, compared to 26% of participants when they were first diagnosed. Vocational achievement was generally more limited, even compared to those with other personality disorders. However, those whose symptoms had remitted were significantly more likely to have good relationships with a romantic partner and at least one parent, good performance at work and school, a sustained work and school history, and good psychosocial functioning overall.

There is a considerable personality disorder diagnostic co-occurrence. Patients who meet the DSM-IV-TR diagnostic criteria for one personality disorder are likely to meet the diagnostic criteria for another. Diagnostic categories provide clear, vivid descriptions of discrete personality types but the personality structure of actual patients might be more accurately described by a constellation of maladaptive personality traits.

"Sites used DSM-III-R criterion sets. Data obtained for purposes of informing the development of the DSM-IV-TR personality disorder diagnostic criteria."

Abbreviations used: "PPD – Paranoid Personality Disorder, SzPD – Schizoid Personality Disorder, StPD – Schizotypal Personality Disorder, ASPD – Antisocial Personality Disorder, BPD – Borderline Personality Disorder, HPD – Histrionic Personality Disorder, NPD – Narcissistic Personality Disorder, AvPD – Avoidant Personality Disorder, DPD – Dependent Personality Disorder, OCPD – Obsessive-Compulsive Personality Disorder, PAPD – Passive-Aggressive Personality Disorder."

About 25-40% of youths diagnosed with conduct disorder qualify for a diagnosis of antisocial personality disorder when they reach adulthood. For those that do not develop ASPD, most still exhibit social dysfunction in adult life.

ASPD commonly coexists with the following conditions:

When combined with alcoholism, people may show frontal function deficits on neuropsychological tests greater than those associated with each condition.

The World Health Organization's (WHO) "International Statistical Classification of Diseases and Related Health Problems", 10th Edition (ICD-10) lists narcissistic personality disorder under "Other specific personality disorders". It is a requirement of ICD-10 that a diagnosis of any specific personality disorder also satisfies a set of general personality disorder criteria.

This diagnostic rubric is not recommended for general use because it is not clearly demarcated either from simple schizophrenia or from schizoid or paranoid personality disorders, or possibly autism and Asperger syndrome as currently diagnosed. If the term is used, three or four of the typical features listed above should have been present, continuously or episodically, for at least 2 years. The individual must never have met criteria for schizophrenia itself. A history of schizophrenia in a first-degree relative gives additional weight to the diagnosis but is not a prerequisite.

Conduct disorder is classified in the fourth edition of "Diagnostic and Statistical Manual of Mental Disorders" (DSM). It is diagnosed based on a prolonged pattern of antisocial behaviour such as serious violation of laws and social norms and rules in people younger than the age of 18. Similar criteria are used in those over the age of 18 for the diagnosis of antisocial personality disorder. No proposed revisions for the main criteria of conduct disorder exist in the "DSM-5"; there is a recommendation by the work group to add an additional specifier for callous and unemotional traits. According to DSM-5 criteria for conduct disorder, there are four categories that could be present in the child's behavior: aggression to people and animals, destruction of property, deceitfulness or theft, and serious violation of rules.

Almost all adolescents who have a substance use disorder have conduct disorder-like traits, but after successful treatment of the substance use disorder, about half of these adolescents no longer display conduct disorder-like symptoms. Therefore, it is important to exclude a substance-induced cause and instead address the substance use disorder prior to making a psychiatric diagnosis of conduct disorder.

NPD has a high rate of comorbidity with other mental disorders. Individuals with NPD are prone to bouts of depression, often meeting criteria for co-occurring depressive disorders. In addition, NPD is associated with bipolar disorder, anorexia, and substance use disorders, especially cocaine. As far as other personality disorders, NPD may be associated with histrionic, borderline, antisocial, and paranoid personality disorders.

Perhaps due to their perceived rarity, the dissociative disorders (including DID) were not initially included in the Structured Clinical Interview for DSM-IV (SCID), which is designed to make psychiatric diagnoses more rigorous and reliable. Instead, shortly after the publication of the initial SCID a freestanding protocol for dissociative disorders (SCID-D) was published. This interview takes about 30 to 90 minutes depending on the subject's experiences. An alternative diagnostic instrument, the Dissociative Disorders Interview Schedule, also exists but the SCID-D is generally considered superior. The Dissociative Disorders Interview Schedule (DDIS) is a highly structured interview that discriminates among various DSM-IV diagnoses. The DDIS can usually be administered in 30–45 minutes.

Other questionnaires include the Dissociative Experiences Scale (DES), Perceptual Alterations Scale, Questionnaire on Experiences of Dissociation, Dissociation Questionnaire, and the Mini-SCIDD. All are strongly intercorrelated and except the Mini-SCIDD, all incorporate absorption, a normal part of personality involving narrowing or broadening of attention. The DES is a simple, quick, and validated questionnaire that has been widely used to screen for dissociative symptoms, with variations for children and adolescents. Tests such as the DES provide a quick method of screening subjects so that the more time-consuming structured clinical interview can be used in the group with high DES scores. Depending on where the cutoff is set, people who would subsequently be diagnosed can be missed. An early recommended cutoff was 15-20. The reliability of the DES in non-clinical samples has been questioned.

The fourth, revised edition of the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR) diagnoses DID according to the diagnostic criteria found in section 300.14 (dissociative disorders). It has also been found difficult to diagnose the disorder in the first place, due to there not being a universal agreement of the definition of dissociation. The criteria require that an adult be recurrently controlled by two or more discrete identities or personality states, accompanied by memory lapses for important information that is not caused by alcohol, drugs or medications and other medical conditions such as complex partial seizures. While otherwise similar, the diagnostic criteria for children also specifies symptoms must not be confused with imaginative play. Diagnosis is normally performed by a clinically trained mental health professional such as a psychiatrist or psychologist through clinical evaluation, interviews with family and friends, and consideration of other ancillary material. Specially designed interviews (such as the SCID-D) and personality assessment tools may be used in the evaluation as well. Since most of the symptoms depend on self-report and are not concrete and observable, there is a degree of subjectivity in making the diagnosis. People are often disinclined to seek treatment, especially since their symptoms may not be taken seriously; thus dissociative disorders have been referred to as "diseases of hiddenness".

The diagnosis has been criticized by supporters of therapy as a cause or the sociocognitive hypothesis as they believe it is a culture-bound and often health care induced condition. The social cues involved in diagnosis may be instrumental in shaping patient behavior or attribution, such that symptoms within one context may be linked to DID, while in another time or place the diagnosis could have been something other than DID. Other researchers disagree and argue that the existence of the condition and its inclusion in the DSM is supported by multiple lines of reliable evidence, with diagnostic criteria allowing it to be clearly discriminated from conditions it is often mistaken for (schizophrenia, borderline personality disorder, and seizure disorder). That a large proportion of cases are diagnosed by specific health care providers, and that symptoms have been created in nonclinical research subjects given appropriate cueing has been suggested as evidence that a small number of clinicians who specialize in DID are responsible for the creation of alters through therapy.

Currently, genetic research for the understanding of the development of personality disorders is severely lacking. However, there are a few possible risk factors currently in discovery. Researchers are currently looking into genetic mechanisms for traits such as aggression, fear and anxiety, which are associated with diagnosed individuals. More research is being conducted into disorder specific mechanisms.

Theodore Millon identified five subtypes of the compulsive personality (2004). Any compulsive personality may exhibit one or more of the following:

Because of reduced levels of trust, there can be challenges in treating PPD. However, psychotherapy, antidepressants, antipsychotics and anti-anxiety medications can play a role when an individual is receptive to intervention.

In the absence of a standardized diagnosis system, many popular, informal classification systems or checklists, outside the DSM and ICD, were created out of clinical and parental experience within the field known as attachment therapy. These lists are unvalidated and critics state they are inaccurate, too broadly defined or applied by unqualified persons. Many are found on the websites of attachment therapists. Common elements of these lists such as lying, lack of remorse or conscience and cruelty do not form part of the diagnostic criteria under either DSM-IV-TR or ICD-10. Many children are being diagnosed with RAD because of behavioral problems that are outside the criteria. There is an emphasis within attachment therapy on aggressive behavior as a symptom of what they describe as attachment disorder whereas mainstream theorists view these behaviors as comorbid, externalizing behaviors requiring appropriate assessment and treatment rather than attachment disorders. However, knowledge of attachment relationships can contribute to the cause, maintenance and treatment of externalizing disorders.

The Randolph Attachment Disorder Questionnaire or RADQ is one of the better known of these checklists and is used by attachment therapists and others. The checklist includes 93 discrete behaviours, many of which either overlap with other disorders, like conduct disorder and oppositional defiant disorder, or are not related to attachment difficulties. Critics assert that it is unvalidated and lacks specificity.

Dependent personality disorder occurs in about 0.6% of the general population. The disorder is diagnosed more often in females than males; however, research suggests that this is largely due to behavioural differences in interviews and self-reporting rather than a difference in prevalence between the sexes. A 2004 twin study suggests a heritability of 0.81 for developing dependent personality disorder. Because of this, there is significant evidence that this disorder runs in families. Children and adolescents with a history of anxiety disorders and physical illnesses are more susceptible to acquiring this disorder.

Diagnosis is based on the self-reported experiences of the person followed by a clinical assessment. Psychiatric assessment includes a psychiatric history and some form of mental status examination. Since some medical and psychiatric conditions mimic the symptoms of DPD, clinicians must differentiate between and rule out the following to establish a precise diagnosis: temporal lobe epilepsy, panic disorder, acute stress disorder, schizophrenia, migraine, drug use, brain tumour or lesion. No laboratory test for depersonalization-derealization disorder currently exists.

The diagnosis of depersonalization disorder can be made with the use of the following interviews and scales:

The Structured Clinical Interview for DSM-IV Dissociative Disorders (SCID-D) is widely used, especially in research settings. This interview takes about 30 minutes to 1.5 hours, depending on individual's experiences.

The Dissociative Experiences Scale (DES) is a simple, quick, self-administered questionnaire that has been widely used to measure dissociative symptoms. It has been used in hundreds of dissociative studies, and can detect depersonalization and derealization experiences.

The Dissociative Disorders Interview Schedule (DDIS) is a highly structured interview which makes DSM-IV diagnoses of somatization disorder, borderline personality disorder and major depressive disorder, as well as all the dissociative disorders. It inquires about positive symptoms of schizophrenia, secondary features of dissociative identity disorder, extrasensory experiences, substance abuse and other items relevant to the dissociative disorders. The DDIS can usually be administered in 30–45 minutes.

The Cambridge Depersonalization Scale (CDS) is a method for determining the severity of depersonalization disorder. It has been proven and accepted as a valid tool for the diagnosis of depersonalization disorder in a clinical setting. It is also used in a clinical setting to differentiate minor episodes of depersonalization from actual symptoms of the disorder. Due to the success of the CDS, a group of Japanese researchers underwent the effort to translate the CDS into the J-CDS or the Japanese Cambridge Depersonalization Scale. Through clinical trials the Japanese research team successfully tested their scale and determined its accuracy. One limitation is that the scale does not allow for the differentiation between past and present episodes of depersonalization. It should also be noted that it may be difficult for the individual to describe the duration of a depersonalization episode, and thus the scale may lack accuracy. The project was conducted in the hope that it would stimulate further scientific investigations into depersonalization disorder.

The World Health Organization's ICD-10 lists dependent personality disorder as " Dependent personality disorder":

It is characterized by at least 4 of the following:

1. Encouraging or allowing others to make most of one's important life decisions;

2. Subordination of one's own needs to those of others on whom one is dependent, and undue compliance with their wishes;

3. Unwillingness to make even reasonable demands on the people one depends on;

4. Feeling uncomfortable or helpless when alone, because of exaggerated fears of inability to care for oneself;

5. Preoccupation with fears of being abandoned by a person with whom one has a close relationship, and of being left to care for oneself;

6. Limited capacity to make everyday decisions without an excessive amount of advice and reassurance from others.

Associated features may include perceiving oneself as helpless, incompetent, and lacking stamina.

Includes:

- Asthenic, inadequate, passive, and self-defeating personality (disorder)

It is a requirement of ICD-10 that a diagnosis of any specific personality disorder also satisfies a set of general personality disorder criteria.

Being a personality disorder, which are usually chronic and long-lasting mental conditions, avoidant personality disorder is not expected to improve with time without treatment. It is a poorly studied personality disorder and in light of prevalence rates, societal costs, and the current state of research, AvPD qualifies as a neglected disorder.

The diagnostic complexities of RAD mean that careful diagnostic evaluation by a trained mental health expert with particular expertise in differential diagnosis is considered essential. Several other disorders, such as conduct disorders, oppositional defiant disorder, anxiety disorders, post traumatic stress disorder and social phobia share many symptoms and are often comorbid with or confused with RAD, leading to over and under diagnosis. RAD can also be confused with neuropsychiatric disorders such as autism, pervasive developmental disorder, childhood schizophrenia and some genetic syndromes. Infants with this disorder can be distinguished from those with organic illness by their rapid physical improvement after hospitalization. Autistic children are likely to be of normal size and weight and often exhibit a degree of intellectual disability. They are unlikely to improve upon being removed from the home.

According to Professor Emily Simonoff of the Institute of Psychiatry, Psychology and Neuroscience, "childhood hyperactivity and conduct disorder showed equally strong prediction of antisocial personality disorder (ASPD) and criminality in early and mid-adult life. Lower IQ and reading problems were most prominent in their relationships with childhood and adolescent antisocial behaviour."