October 2006, Vol 28, No. 10
Discussion Paper

I never see mental health problems in my practice .... do I?*

Carol P Herbert

HK Pract 2006;28:429-437


This paper discusses selected hard-to-diagnose mental health problems from the author's 30-year experience in family practice, in the context of the high prevalence of mental health problems in ambulatory practice and the importance of intervention by family physicians who may be the only health professional consulted by these patients. A Canadian collaborative model for shared-care between mental health professionals and family doctors is suggested as a possible approach to help busy family doctors in Hong Kong manage complex mental health problems.



* Based on a presentation to the Hong Kong College of Family Physicians, on 5 July, 2006 during the author's tenure as Hong Kong College of Family Physicians Visiting Professor of Family Medicine at The University of Hong Kong.


Mental health problems, including mood disorders related to physical illness, result in a substantial burden of illness worldwide. A significant number of visits to family doctors are related to mental health problems which may or may not be recognized by the patient, or the physician, as a reason for the encounter.

In this paper I will comment on the high incidence and prevalence of mental health problems in ambulatory practice and the importance of appropriate diagnosis and management by family physicians, discuss selected hard-to-diagnose mental health problems from my 30 years of experience in Canadian family practice as examples of conditions that may challenge the diagnostic acumen of the family doctor, and describe a Canadian collaborative model for shared care between mental health professionals and family doctors that could be considered for Hong Kong, to make it easier for busy family doctors to manage complex mental health problems.

Incidence and prevalence of mental health problems in ambulatory care

In community surveys in the United States of America,1 26.2% of people over the age of 18 have a diagnosable mental disorder, with 6% suffering from serious mental disorders. Ambulatory utilization data for 20042 revealed that 48 million visits to office-based physicians were for mental disorders. Additionally, 3.7 million visits to emergency departments and 51.7 million visits to ambulatory care facilities were for mental health problems.

The highest number of ambulatory visits in the United States 2004 Ambulatory Care Survey2 were for depression (21 million); schizophrenia and other psychoses (8.5 million); anxiety (6.2 million); problems related to drugs or alcohol (2.8 million); and attention deficit disorder (5.4 million).

While the prevalence of psychological problems in ambulatory practice in Hong Kong is not known, general population surveys of depression in the elderly range from 0.4% to 19%.3 We can hypothesize that, in accord with the US statistics, up to 25% of Hong Kong patients seen by family doctors may have a diagnosable mental health problem. The rates in Hong Kong may be skewed in private and public practice, given the different proportions of acute and chronic illness seen in the different settings, i.e., it is possible that higher rates of mental health problems will be seen in public practice settings where a higher proportion of patients with chronic illness attend.

A Canadian study4 reports that 72% of persons with mental illness undetected by their doctors had received no treatment for their illness over the course of a year, with 81% of these people reporting that the only doctor they visit is their family physician.4 This reinforces the fact that family doctors have both the opportunity and obligation to diagnose mental health problems that could be helped by treatment since they may be the only health professional seen by the patient.

A study in Taiwan conducted by Liu and colleagues5 supports the finding of under-diagnosis of mental disorders by general practitioners. In a comparison between checklists completed by doctors immediately after an office visit and a Chinese Health Questionnaire with a subsequent structured interview, more than 85% of diagnoses were missed by the doctors (n=990). It is instructive to note that correct diagnosis was more likely for patients with higher socioeconomic status, for patients who did not have a physical illness, when the mental health problem was presented by the patient, for more serious mental disorders and when the illness was of longer duration.

Why is it important to diagnose undetected mental illness? Patients with mental illness have decreased functional abilities,6 increased morbidity and mortality,7 and increased health care costs.8 When significant mental illness is detected, consultation may make a positive difference. Several US studies have shown that significant mental health problems treated by primary care practitioners without psychiatric consultation and advice on management have poorer outcomes9-12 though this does not diminish the important role of the family doctor in mental health care, for detection and ongoing management.

'Hidden' mental health problems

Many patients in primary care who have undetected mental health diagnoses do not present to their doctors with mental health complaints. Moreover, in the early stages of illness, symptoms may be non-specific or general - it may be very difficult to specify a diagnosis. Other patients who have long-standing issues that relate to their social situation or lifestyle, such as drug and alcohol problems or family violence, may be reluctant to disclose to the doctor.

The following case examples come from the author's experience over 30 years of practice, as difficult-to-diagnose mental health problems where the family doctor can play an important role in diagnosis and management:

Borderline personality disorder

Case example: Barbara is a 36-year old woman who has not worked for several years. She has multiple somatic symptoms and is never satisfied with the diagnosis or the intervention. Her usual response to a recommendation is "Yes, but...", followed shortly thereafter by "something must be done, doctor..." regarding whatever is her current symptom. She has seen many doctors, goes to the A&E frequently, and has had multiple hospital admissions for anxiety, depression, and suicidal gestures.

Among the most difficult patients are those with personality disorders - deep-rooted patterns of attitudes and behaviours that interfere with the person's ability to lead their lives and that make the health care relationship difficult. In my experience, the most difficult personality disorder in family practice is the borderline personality disorder.

A small number of borderline patients can consume a lot of doctor's time and health care resources. It is important to recognize the diagnosis in order to establish a treatment plan.

These individuals have had a difficult time with relationships in their lives. They tend to present as 'victims', hard done by in their families, their workplace, and their relationships. Initially, the doctor may see himself as 'rescuer'. "You are the only doctor who has ever listened to me....", but after failing repeatedly in his attempts to 'cure' the patient, the doctor may begin himself to feel 'victimized' by the patient. Unfortunately, the normal reaction to being victimized is to fight back, to 'persecute' the one who is causing you pain. Thus the doctor may become angry and resentful, and reject the patient, recapitulating what has happened repeatedly to the patient in relationships. The rescuer- victim-persecutor triangle is useful to recognize as it can help the doctor to identify the patient's diagnosis by observing his own feelings.

The doctor should avoid over-investigation of symptoms and unnecessary referrals or hospitalizations that can result in a cascade of tests and even surgery. By providing regularly scheduled short visits and reinforcing positive behaviour, e.g., heaping praise on the patient for beginning a walking programme, and ignoring negative behaviours, e.g., complaints of weight gain or chronic pain, it is possible to help such patients gain some self-control. The doctor should resist unscheduled emergency office visits and direct the patient to return for their regular appointment. It is difficult but important not to over-react to suicidal threats which are attention-seeking behaviours.

Consultation with a psychiatrist can be very helpful, as can a shared care arrangement with a mental health clinic. However, it is critical that communication be clear between care-providers to avoid 'splitting', another behaviour that is commonly encountered with these patients who will criticize one health professional to another.

A particularly troublesome behaviour that may be seen is self-injury, usually cutting or burning. While this behaviour cannot be prevented by admonishment, the doctor can respond to the identification of the injury by asking the patient what they are experiencing before and during the behaviour; in other words, what is being achieved by the self-injury. He can ask if there are other ways that the patient could relieve the pressure they feel building up inside them. If a patient begins to cut or burn themselves during the office visit, the doctor can refuse to continue the consultation unless they stop the behaviour.

Family violence

Case example: 8-year old child with vaginal discharge was found to have gonorrhea and reported to Child Protection Services. A cursory investigation was done but the child denied any sexual abuse. According to the prevailing medical opinion in 1980, it was decided that she had contracted the disease from the bedsheets in the parental bed. Four years later the child ran away from home to live on the street. Shortly after, her younger sister reported sexual abuse by the step-father that had been occurring over the past several years, involving both children.

Case example: 30-year old woman with two children who came to the office requesting an anxiolytic. In response to questioning, she acknowledged regular beatings by her husband but stated that she could not leave the home as she had no money and limited education. She stated that she had a long-term plan, and was currently completing her secondary school and saving money. She denied that there was any violence towards the children. We created an emergency 'escape plan' in case of escalation of the violence. I saw her for regular visits and prescribed small amounts of anxiolytic medication intermittently over the next two years until she finally felt able to leave the relationship.

Individuals who are victims of physical, sexual or emotional abuse may present with what appear to be unrelated problems, though our understanding of child sexual abuse has certainly improved over the past two decades. Conduct disorder in adolescents and depression in women are common psychiatric diagnoses that may mask an underlying problem of violence in the family. Failure to identify the underlying cause results in mis-diagnosis and potential long-term medication for the woman with no intervention for the perpetrator.

Another common presentation of physical abuse is unexplained physical injury in women and children. Either from fear or from a desire to protect the perpetrator and keep the family intact, the patient may deny abuse. The doctor needs to look for patterns - injuries that do not fit the history; evidence of blows to the body that spare the exposed face and limbs; shaking injuries in infants. Risk factors for child abuse include premature birth and colic in infants, parental alcohol or drug abuse, and difficult family situations, including poverty, job loss, single mothers, or young parents. However, child abuse occurs across all social classes, as do all forms of family violence.

Sexually abused children may be anxious or depressed or have sleep problems, or they may act out sexually. But often there are few clues. Physical evidence is often absent, however evidence of absent hymen, gaping anus, bruising around the genitals, or vaginal discharge demands further investigation. If the doctor has reason to believe that abuse may be occurring, he or she is bound to report to Child Protective Services.

Pregnancy is an independent risk factor for woman abuse. First episodes of violence may occur during pregnancy or an existing abusive relationship may be exacerbated. Increasing severity of physical violence is common and may lead to life-threatening injuries or death. Thus identification of woman abuse may be lifesaving.

Unlike child abuse where doctors have a legal obligation to report to Child Protective Services, in woman abuse the doctor has a moral responsibility to counsel the woman about the risk of repetitive and escalating violence and to develop a safety plan, i.e., identification of a safe house to which the woman can flee, preparation of money and important documents for rapid retrieval. Referral to counselling can help a woman to determine her options.

Eating disorders in children and adolescents

Case example: A 15-year old competitive figure-skater has been losing weight. She is brought to see you by her parents. They report that she has developed odd eating rituals, including not letting the food touch her lips as she puts food into her mouth and separating the foods on her plate, becoming very anxious if the foods touch one another. In response to questioning, you determine that menarche was at age 13, but that her menses stopped several months ago.

Eating disorders (ED) in adolescents are common, with an incidence of up to 5%, making ED the third most common chronic illness in adolescent girls. Eating disorders are a spectrum that includes anorexia, bulimia, and EDNOS - eating disorders not otherwise diagnosed, perhaps the more common ED diagnosis in primary care where the illness may be detected in its undifferentiated state. High risk children include competitive figure-skaters, gymnasts, and dancers, particularly as they enter menarche and their bodies start to mature. Children under pressure to excel at school may also be at higher risk.

Some children present with physical symptoms and signs of weight control behaviours, including self-induced vomiting, chronic diarrhoea, marks on the knuckles of the hands from forced vomiting, and signs of malnutrition. Potentially irreversible changes include growth retardation, loss of dental enamel, pubertal delay or arrest, impaired bone mass that predisposes to osteoporosis in later life, and structural brain changes.13,14

However, the challenge for the doctor is to diagnose early symptoms before the illness is fully developed in order to prevent escalation. The threshold for intervention should be low, even if the child denies there is an eating problem. Parents and classmates who report ritualistic eating patterns should be taken seriously.

While many cases can be managed effectively in primary care practice by providing parental advice and support and monitoring the child, it is useful to have nutritional and mental health consultation. Family therapy may help to break up patterns of behaviour and communication that have contributed to the ED and/or to deal with management issues that affect the whole family.

Hospitalization may be necessary for severe malnutrition, physiologic instability, severe mood disorder, or failure of outpatient treatment. Costs of care may be high since full-blown ED tend to be chronic and recurrent. Hence early intervention may be cost-effective.

ED do appear to be on the rise with weight issues appearing in children as young as 6 or 7 years. As doctors deal with prevention of obesity in children, it is important to ensure that they do not promote anorexia and bulimia by making children embarrassed and ashamed that they are fat. Focusing on increased physical exercise is preferable to diets in overweight children; caregivers can emphasize health and strength, rather than weight.

Parents should be made aware that there are now websites promoting ED ('pro-ana' and 'pro-mia') that their children may be accessing and from which they can learn negative eating and weight control behaviours.

Mild cognitive impairment

Case example: An older couple presents for monitoring of her diabetes and his high blood pressure. The wife states that her 70-year old husband has been very forgetful of names over the past several months. They are worried that this may be early dementia.

Many older adults notice memory changes and may complain to the doctor about inability to remember names, in particular. Some of these individuals may suffer from Mild Cognitive Impairment (MCI).15 This condition is characterized by subjective complaints of cognitive impairment with some objective evidence on standardized cognitive testing such as the Folstein Mini-Mental Status (MMS) test, which can be administered in the office or in a geriatric assessment setting. There is no significant impairment in activities of daily living (ADL), which distinguishes this condition from early dementia.

Individuals with MCI have a 10-15% risk of developing dementia as opposed to 1-2% in the general population over 65.15 While there is still debate over whether therapeutic intervention can slow the progression of dementia, individuals may wish to have neurologic assessment and testing to determine their risk status and to make informed choice about drugs such as donepezil which show some evidence of improved MMS scores though clinically significant improvement has not been proven.15

Atypical depression in older adults

Case example: A 65-year old woman who has had severe rheumatoid arthritis for many years, stops taking her medication. She stays in bed, though there has been no apparent change in her medical condition. She does not recognize some family members who come to visit her.

Depression in older adults is common, under-diagnosed and under-treated worldwide, including in Hong Kong.3 Symptoms may be under-reported by patients who 'suffer in silence' or attributed by them or their doctors to their physical state. Clues to atypical depression include anxiety and worry that appears or worsens without evident cause; somatization; associated physical illness; complaints of memory loss; pseudo-dementia; expressed hopelessness; lack of adherence to treatment; or a change in general functioning, not otherwise explained, e.g., someone who has been previously active takes to their bed or avoids family and friends. Risk factors include functional impairment associated with age and illness; chronic illness; medications that may predispose to depression; and psychosocial factors.16

What is important to note for the doctor is that these patients respond to treatment. Effective antidepressant treatment or sometimes ECT will reverse a pseudo-dementia state or prevent active or passive (by stopping medications or not eating) suicide.

Memory loss is a particularly difficult symptom to sort out since it can represent normal ageing, MCI, early dementia, or depression. If in doubt as to a treatable cause, a geriatrician may be of help to sort out the situation so that ongoing care can be maintained by the family doctor.

Post-traumatic stress disorder (PTSD)

Case example: Joan is a 30-year old woman who complains of insomnia and episodic shortness of breath. The doctor determines that she is having panic attacks. On questioning, she reports that she has memory blocks for some of her past, particularly for her late adolescent years.

PTSD is another condition that is difficult to recognize in practice and therefore under-diagnosed and under-treated. Symptoms may not be declared by the patient or they may be non-specific. Memory loss, somatization, anxiety and worry, and depression may be clues that should prompt the physician to inquire about possible precipitating events. Risk factors include history of sexual abuse or sexual assault, other major traumatic events, such as fire, natural disaster, or physical violence.

Recent evidence shows changes in the brain of individuals with PTSD, supporting evidence that this is a 'real' condition.17,18 The majority of patients treated with psychotherapy in randomized trials recover or improve. However, the majority of patients post-treatment continue to have residual symptoms and, in an unselected population within the community, the effectiveness of standard psychotherapy approaches is not known.19

Collaborative mental health care

Family doctors have the responsibility - and the ability - to diagnose mental health problems. They are particularly well placed to intervene early, even before a formal diagnosis is evident, e.g., in the early stages of an eating disorder. Sometimes the management approach is clear. Some doctors may have taken additional training in psychological medicine that provides them with additional skills, both for diagnosis and treatment.

However, as stated earlier in this paper, it appears that serious mental disorders are best managed by family doctors who have the benefit of consultation with a psychiatrist. In some centres, formal collaborative mental health programmes of care have been established to assist community-based doctors to obtain rapid consultation, to move patients through the mental health system appropriately if they need more intensive care, and to 'step-down' patients back to the community after hospitalization. Such programmes do not 'take' patients from the family doctor, but rather assist the practitioner to formulate a treatment plan and to manage complex illnesses with the assurance that they have support available if problems arise.

A working definition for Collaborative Mental Health Care (CMHC) has been developed by the Collaborative Working Group on Shared Mental Health Care, a shared venture of the Canadian Psychiatric Association and the College of Family Physicians of Canada, as follows: the process of collaboration between a family physician and mental health professional that enables responsibilities for care to be apportioned according to the treatment needs of the patient and the respective skills of the mental health professional and the family physician.

In this model, the family doctor continues to see cases, but has regular contact with a mental health clinician. The family doctor collaboratively plans for changes in treatment, prescribes for the patient and provides ongoing care for all health conditions, including the mental health problem. The mental health professional may be housed in a separate facility or may spend time on site in a community practice setting, e.g., a Family Medicine Unit, where they may see patients in consultation, observe the doctor in the office visit with the patient and provide advice, or provide advice based on chart reviews.

Another useful model is the Transition into Primary Care Psychiatry or TIPP model that is in place in London, Ontario, Canada. As shown in Figure 1, this is a patient-centred model whereby resources are provided to patients in the community. The model allows for the return of patients who have been acutely ill in hospital to their community-based doctor, who continues to have access to a range of resources as needed.

Collaborative mental health care has been well documented in a series of research papers available from the website www.ccmhi.ca, products of the Canadian Collaborative Mental Health Initiative. This series includes a paper that reviews the evidence for better practices in collaborative mental health care.20 Experimental evidence suggests eleven conclusions:

  1. Collaborative relationships between primary care physicians and other mental health care providers do not happen instantly or with out work. They require preparation, time, and supportive structures. System-level collaboration also requires preparation, service reorganization and time to develop.
  2. Co-location is important for both providers and patients. Linking professionals works best when providers know each other and especially if they are physically close to one another. Similarly, patients seen in familiar non-stigmatizing locations like their family physician's office do better.
  3. Degree of collaboration does not in itself appear to predict clinical outcome. Even limited collaboration has positive outcomes.
  4. The pairing of collaboration with treatment guidelines appears to offer important benefits over either intervention alone in patients with depressive disorders.
  5. Collaboration paired with treatment guidelines for depression may have a differential effect on outcome, with patients with more severe disorder responding better. At present, there is more evidence to support targeting collaborative interventions at major depressive disorders.
  6. One of the most powerful predictors of positive clinical outcomes in studies of collaborative care for depression was the inclusion of follow-up as part of the study protocol.
  7. Efforts to increase medication adherence through collaboration with other health care professionals were also a common component of many successful studies. However, there is no clear relationship between medication adherence and clinical outcome.
  8. Collaboration alone has not been shown to produce skill transfer or enduring change in primary care physician knowledge or behaviours in the treatment of depression. Service re-structuring is essential to support changes in practice patterns.
  9. Enhanced patient education, usually by a health professional other than the primary care physician, was a component of many of the studies with good outcomes.
  10. Collaborative interventions established for research may be difficult to sustain once the funding for the study ends. It is important to build in sustainability, e.g., by engaging permanent staff in the project.
  11. Patient choice about treatment modality may be an important factor in treatment engagement, specifically the option of psychotherapy versus medication.

In addition, the same website allows access to twelve toolkits to support the implementation of collaborative mental health care, an approach that will assist health planners, providers, educators, consumers, families and caregivers, to provide the best possible management for mental health problems.


Given that family doctors in Hong Kong are faced with large numbers of patients each day, the feasibility and practicality of a shared-care model for complex mental health issues should be considered. With such a model, complex patients could be referred readily to other health care providers for development of a treatment plan, acute intervention, or special services such as nutritional counselling, with the assurance that they would be returned to the family doctor for ongoing management. This approach may encourage doctors to take on management of conditions that they might otherwise avoid.


The author acknowledges the advice and references provided by colleagues from the Department of Psychiatry at the University of Western Ontario, particularly Drs. Sandra Fisman, Jatinder Takhar, and David Haslam.

Key messages

  1. Mental health problems are common in family practice, though they may be under-reported by patients and unrecognized by physicians.
  2. Family doctors have both opportunity and obligation to diagnose and manage mental health problems as they may be the only health professional consulted.
  3. Case presentations may be confusing, with non-specific symptoms and signs in a range of conditions.
  4. Collaborative care between mental health professionals and family doctors may make it easier for busy Hong Kong family doctors to manage complex mental health problems

Carol P Herbert, MD, CCFP, FCFP, FRCP(Glasg)
Schulich School of Medicine & Dentistry, The University of Western Ontario.

Correspondence to: Professor Carol P Herbert, Schulich School of Medicine & Dentistry, The University of Western Ontario, London, Ontario, Canada N6A 5C1.

  1. http:www.nimh.nih.gov/HealthInformation/statisticsmenu.cfm. (accessed 23 Oct, 2006).
  2. http:www.cdc.gov/nchs/FASTATS/mental.htm. (accessed 23 Oct, 2006).
  3. Lam CLK, Lee PWH, Fong DYT, et al. Unrecognized Psychological problems are Common and Important in Chinese Elderly Patients in Primary Care, unpublished paper.
  4. Parikh SV, Lin E, Lesage AD. Mental health treatment in Ontario: selected comparisons between the primary care and specialty sectors. Can J Psychiatry 1997; 42:929-934.
  5. Liu S, Mann A, Cheng A, et al. Identification of common mental disorders by general medical doctors in Taiwan. General Hosp Psychiatry 2004;26:282-288.
  6. Wells KB, Stewart A, Hays RD, et al. The functioning and well-being of depressed patients. Results from the Medical Outcomes Study. JAMA 1989;262:914-919.
  7. Osborn DPJ. The poor physical health of people with mental illness. West J Med 2001; 175:327-332.
  8. Simon G, Ormel J, VonKorff M, et al. Health care costs associated with depressive and anxiety disorders in primary care. Am J Psychiatry 1995;152:352-357.
  9. Smith GR Jr, Rost KT, Kashner TM. A trial of the effect of a standardized psychiatric consultation on health outcomes and costs in somatizing patients. Arch Gen Psychiatry 1995;52:238-243.
  10. Lin EHB, Von Korff M, Russo J, et al. Can Depression Treatment in Primary Care Reduce Disability?: A Stepped Care Approach. Arch Fam Med 2000;9:1052-1058.
  11. Katon W, Von Korff M, Lin E, et al. Collaborative management to achieve treatment guidelines. Impact on depression in primary care. JAMA 1995;273:1026-1031.
  12. Katon W, Von Korff M, Lin E, et al. Stepped Collaborative Care for Primary Care Patients With Persistent Symptoms of Depression: A Randomized Trial. Arch Gen Psychiatry 1999;56:1109-1115.
  13. American Academy of Pediatrics Committee on Adolescence. Policy Statement: Identifying and Treating Eating Disorders. Pediatrics 2003;111:204-211. (http://www.pediatrics.org/cgi/content/full/111/1/204 )
  14. Golden NH, Katzman DK, Kreipe RE, et al. Eating Disorders in Adolescents: Position Paper of the Society for Adolescent Medicine. J Adolescent Health 2003; 33:496-503.
  15. Frank AR, Petersen RC. An Office Diagnosis of Mild Cognitive Impairment. Geriatrics & Aging 2006; 4:270-274.
  16. Sylvius JL. Atypical Presentations of Depression. Geriatrics & Aging 2006; 9:4:246-249.
  17. Lanius RA, Williamson PC, Densmore M, et al. The Nature of Traumatic Memories: A 4-T fMRI Functional Connectivity Analysis. Am J Psychiatry 2004;160:1-9.
  18. Lanius RA, Williamson PC, Bluhm TL, et al. Functional Connectivity of Dissociative Responses in Posttraumatic Stress Disorder: A Functional Mangentic Resonance Imaging Investigation. Biol Psychiatry 2005;57: 873-884.
  19. Bradley R, Greene J, Russ E, et al. A Multidimensional Meta-Analysis of Psychotherapy for PTSD. Am J Psychiatry 2005;162:214-227.
  20. Craven M, Bland R. Better Practices in Collaborative Mental Health Care: An Analysis of the Evidence Base. Can J Psych 2006; 51 (Suppl 1, May). Available at www.ccmhi.ca. (accessed 12 June, 2006).