Understanding the expanding role of primary care physicians (PCPs) to primary psychiatric care physicians (PPCPs): enhancing the assessment and treatment of psychiatric conditions

Abstract

Aim In the current healthcare system primary care physicians (PCPs) have, in effect, become the primary psychiatric care physicians (PPCPs) for many of their patients. Being the PPCP in an already busy and stressful medical industry presents additional time management and treatment challenges to successfully manage patients' medical and psychiatric needs. The aim of the study was to ascertain PCPs' psychiatric assessment and treatment practices and to determine the extent to which PCPs have a need for using a structured psychiatric assessment tool.

Method We sent 300 PCPs a survey to examine their psychiatric assessment and treatment practices. A one‐page questionnaire was used to inquire about PCPs' psychiatric care practice habits including types of conditions treated, psychiatric medications prescribed, assessment methods used, interest in using a structured assessment tool and referral sources used. Sixty‐eight usable surveys (23%) were returned.

Results PCPs identify approximately one‐third of their patients as mental health patients. They are treating a wide range of psychiatric conditions and prescribing a variety of psychiatric medications. The vast majority are using traditional clinical interviewing as their primary method of psychiatric assessment. However, the majority were willing to use a structured psychiatric assessment tool.

Conclusion PCPs are serving a useful role in providing psychiatric treatment to many of their patients. Using a more structured psychiatric assessment method in practice could ultimately strengthen the assessment and treatment of psychiatric conditions in primary care settings.

Introduction

Research indicates that in recent years the delivery of mental health services has changed significantly.^1–3 Increasing numbers of patients, especially depressed patients, are receiving psychiatric treatment from primary care physicians (PCPs), rather than from mental health specialists. In general, PCPs are more likely to provide psychiatric care rather than refer their patients to mental health specialists.^3–5 Research has also shown that over a ten‐year period from 1987 to 1997, the percentage of patients who received psychiatric medication from PCPs increased from 37.3% to 74.5%.⁶ These trends demonstrate that the role of the primary care physician has expanded such that PCPs are also becoming the primary psychiatric care physician (PPCP) for a considerable number of their patients.

The expanding role of primary care physicians may be due to a variety of reasons. For example, some research shows that up to two‐thirds of PCPs may be functioning as PPCPs due to shortages of mental health care providers, limited insurance to cover psychiatric treatment,⁷ or possibly due to patients' reluctance to access available psychiatric services due to the ‘stigma’ associated with seeing a psychiatrist. Additionally, with the development of safer psychotropic medications (i.e. with safer side effect profiles) such as SSRIs, PCPs may feel more comfortable with treating psychiatric disorders. Moreover, the increased tendency within psychiatry and society as a whole to conceptualise psychiatric disorders as ‘medical problems’ requiring medical treatment (i.e. medication) has expanded the ‘clinical purview’ of primary care to provide psychiatric care. Finally, with easier public access to electronic information (e.g. the internet) and greater advertising freedom (e.g. direct‐to‐consumer advertising), people have become more knowledgeable about psychiatric conditions and treatments and, as a result, are more willing to seek psychiatric treatment. This, in turn, has increased physicians' willingness to provide psychiatric treatment to their patients.

For example, in a study of pseudo‐patients who presented with depression, PCPs were willing to prescribe psychiatric medication to the majority of these patients. In particular, PCPs were significantly more likely to prescribe a psychiatric medication if the patient requested a specific brand of medication or if the patient requested medication in general.⁸ The extent to which PCPs' prescribing practices, in general, were influenced by patients' requests for medications is unknown. PCPs may very well have intended to promote the mental health of patients by responding to a real need they perceived in the patient independent of the patient's request for a medication. However, the study also raises the possibility that PCPs may be willing to prescribe psychiatric medication after a very brief ‘psychiatric interview’ with the patient.

Despite improvements in medication treatments and society's greater acceptance of psychiatric conditions as medical conditions, research suggests that, in general, psychiatric care in primary care is still beset with its own problems such as overuse of medi‐cations,⁹ failure to diagnose common disorders like depression and anxiety,^10,11 failure to properly identify co‐morbid conditions,^12,13 underdiagnosis^14–17 and possible misdiagnosis and mistreatment (e.g. treating bipolar depression as a major depressive disorder).¹⁸ Problems with primary care adequately addressing the psychiatric needs of patients has been identified in the US Surgeon General's Mental Health Report,¹⁹ which noted that primary care has revealed low rates of recognising and treating depression and that ‘poor recognition leads to unnecessary and expensive diagnostic procedures, particularly in response to patients' vague somatic complaints’ (p. 269).

However, the US Surgeon General¹⁹ reports that ‘primary care offers the potential advantages of proximity, affordability, convenience, and coordination of care for mental and somatic disorders and many older people prefer to receive mental health treatment in primary care’ (p. 372). Similarly, the World Health Organization (WHO) in conjunction with the World Organization of Family Doctors (Wonca), supports integrating mental health into primary care and has formally presented and reported on the justification and advantages of providing mental health treatment in primary care.²⁰

Psychiatric assessment in primary care

The expanding role of PCPs in this new era of mental health treatment has prompted researchers to explore ways of improving PCPs' training in diagnosing and treating psychiatric disorders.^17,21,22 In particular, research has indicated, for example, the importance of educating primary care physicians about screening, treating and managing patients with bipolar disorder.¹⁸ Others have indicated that over‐reliance on the traditional unstructured clinical interview can cause a health provider to under‐recognise co‐morbid conditions,¹³ and that making efforts to gather as much information as possible can help reduce inaccurate diagnosis.¹² However, research has also shown that in primary care there is no agreed method for assessing or documenting a patient's psychiatric symptoms.¹⁷ In fact, the US Surgeon General¹⁹ noted that, as it currently stands, primary care ‘has limited capacity to identify patients with common mental disorders and to provide the proactive follow‐up that is required to retain patients in treatment’ (p. 373).

The need to improve psychiatric assessment in primary care has motivated some researchers to develop brief screening tools for common psychiatric conditions seen in general practice. For example, some researchers have developed a brief psychiatric screening tool by using 12 of the 34 items from the Somatic and Psychological Health Report (SPHERE) questionnaire.²³ Various other psychiatric screening tools for primary care settings have been developed, such as; the Primary Care Evaluation of Mental Disorders (PRIME‐MD); the Patient Health Questionnaire (PHQ); the Rand Mental Health Inventory (MHI); and the DUKE‐AD Scale.²⁴ PRIME‐MD, which seemed to have the potential to serve as a practical and useful screening tool in primary care settings (it takes about eight minutes to complete), faced problems in being successfully incorporated into practice, possibly due to the need for training, and ultimately has not been used in any consistent way in the average primary care setting.²⁵

More recently, the WHO and Wonca²⁰ are promoting the use of the Global Mental Health Assessment Tool (GMHAT/PC), a computerised psychiatric assessment tool which has been translated into several languages, as a way to better integrate mental health assessment into primary care. Recognising the need for a universal method of psychiatric assessment in primary care is an important step in enabling and enhancing PCPs' ability to more accurately assess and treat their patients' psychiatric needs. However, using this tool routinely in primary care clinical practice can require considerable time (12–15 minutes per patient),²⁶ resources and training.

In general, research shows that patients do not mind completing brief questionnaires and that the majority of family physicians have used at least one of the following measures: the Mini‐Mental State Examination (cognitive functioning) (67.3%); the Beck Depression Inventory (37%); the Zung Self‐Rating Depression Scale (35%); the Conners Rating Scales (attention deficit hyperactivity disorder (ADHD)) (30.7%); the Hamilton Rating Scale for Anxiety (18.6%); the Geriatric Depression Scale (17.2%); the Hamilton Rating Scale for Depression (15.3%); the Primary Care Evaluation of Mental Disorders (mood, anxiety, somatoform disorders, alcohol abuse/dependence, bulimia nervosa) (9.7%); the Global Deterioration Scale (dementia) (3.9%); and the Yale–Brown Obsessive Compulsive Scale (3.5%).²⁷

Enhancing the psychiatric assessment methods of PCPs could serve a useful purpose in enabling PCPs to gain a more thorough and accurate assessment of their patients' psychiatric status and ultimately improve treatment practices. Using a more structured assessment method would also enable PCPs to justify their treatment practices and to more reliably assess their patients' response to treatment. Furthermore, enhancing PCPs' psychiatric assessment practices would strengthen PCPs' ‘skill mix’ and enable them to function more effectively as their patients' PPCP.

Although PCPs routinely treat psychiatric conditions, it is unclear to what extent today's PCPs use structured assessment methods, and the extent to which they coordinate adjunctive mental health treatment. To further examine these issues, we conducted a survey of Rhode Island primary care physicians. In particular, we examined PCPs' psychiatric assessment methods, the types of psychiatric conditions they treated, the psychiatric medications they prescribed and the extent to which PCPs coordinated adjunctive psychiatric services (i.e. psychotherapy, referrals to psychiatrists) for their patients' care.

Method

Three‐hundred PCPs were randomly selected from the Directory of Board Certified Internists obtained from the Rhode Island Department of Health. Subjects were mailed a questionnaire containing various questions about their psychiatric assessment and treatment practices (see Appendix 1 for a copy of the questionnaire). The questionnaire items inquired about: 1) the percentage of mental health patients on the PCP's caseload; 2) whether the PCP used any psychiatric screening tool and, if not, whether the PCP was willing to use a screening tool; 3) the ranking, in order of frequency, of psychiatric conditions treated; 4) classes of psychiatric medications prescribed and 5) the percentage of their mental health patients for whom they a) prescribed medication only, b) referred for therapy only, c) prescribed medication and referred for therapy or d) referred to a psychiatrist.

The 300 PCPs were sent a cover letter explaining the purpose of the study, a confidentiality statement and a questionnaire. Within two months of the initial mailing, 48 of the PCPs returned completed questionnaires. Shortly after, a second full mailing was sent to all 300 subjects inviting each subject to return the survey if they had not already done so. Following the two mailings, 81 questionnaires (27% of the sample) were eventually returned. Some of the questionnaires were unsuitable for analysis for reasons such as retirement or insufficient data. Ultimately, 68 usable questionnaires (23% of the sample) were available for analysis. The majority (58) of these usable questionnaires were from PCPs. However, some questionnaires (10) were from specialists such as pulmonologists, gastroenterologists, oncologists and hospitalists who indicated that they provided psychiatric treatment to some of their patients.

Results

The PCP sample consisted of 50 Doctors of Medicine (MDs) and three Doctors of Osteopathy (DOs); five did not indicate a degree. The average number of years of practice were 11.4 with a range of 2–37 years. Thirty‐seven were males, 17 were females, and four failed to report their gender. The age range for PCPs varied across the following: 25–35 years old (12.3%, n = 7), 36–45 years old (31.6%, n = 18), 46–55 years old (33.3%, n = 20), 56–65 years old (12.3%, n = 7), 66–75 years old (1.8%, n = 1): 8.8% (n = 5) did not respond (for the 76+ age category, we did not have anyone who returned a usable questionnaire).

The average number of patients on a PCP's caseload was 3461, and on average PCPs identified 30.3% of their patients as mental health patients. This suggests that each year, almost one‐third of a PCP's patients (i.e. over 1000 patients) are being treated for a mental health problem. This finding is consistent with past research suggesting that approximately 30% of a PCP's patients are treated for a psychiatric condition.^21,28 Our findings also indicated that approximately 15% of the PCPs reported that at least 50% of their patients were psychiatric patients and approximately 6% of the PCPs identified 70% of their patients as mental health patients. Interestingly, 50% of the non‐PCPs (medical specialists) indicated that they were treating at least 50% of their patients for a psychiatric condition (see Table 1 for a summary of the PCPs' questionnaire responses; Table 2 includes the combined responses for PCPs and non‐PCPs).

Table 1.

Demographics and practice characteristics of PCPs

	n	%
Gender
Male	37	63.8
Female	17	29.3
No gender given	4	6.9
Degree and practice
MD	50	86.2
DO	3	5.1
Degree not given	5	8.6
Average years of practice		11.42
Average percentage of patients receiving
Just medication		40
Just therapy		13
Medication and therapy		30
Psychiatric referral		17
PCP medication prescribing
Antidepressants	58	100
Anti‐anxieties	53	91.3
Sleep aids	51	87.9
Cognitive enhancers	45	77.6
Stimulants	25	43.1
Antipsychotics	24	41.3
Mood stabilisers	21	36.2

Conditions treated
	Considered 1	2	3
Depressive disorders	25	13	3
Anxiety disorders	15	27	2
Panic attacks	0	1	1
Stress disorders	3	6	9
Bipolar disorders	1	0	2
ADHD	1	0	2
Substance abuse	0	0	5

Screening tool
Using a current screening tool		Interested in using a new brief psychiatric tool
Yes	13(22.4%)	Yes	37(63.8%)
No	45(77.5%)	No	17(29.3%)
		Unsure	4(6.9%)

Table 2.

Demographics and practice characteristics of PCP and non‐PCP physicians

	n	%
Gender
Male	42	61.8
Female	21	30.8
No gender given	5	7.4
Degree and practice
MD	58	85.2
DO	3	4.4
Degree not given	7	10.2
Average years of practice		11.57
Average percentage of patients receiving
Just medication		38
Just therapy		13
Medication and therapy		28
Psychiatric referral		22
Physician medication prescribing
Antidepressants	67	98.5
Anti‐anxieties	62	91.1
Sleep aids	60	88.2
Cognitive enhancers	45	66.1
Stimulants	26	38.2
Antipsychotics	26	38.2
Mood stabilisers	22	32.3

Conditions treated
	Considered 1	2	3
Depressive disorders	30	15	3
Anxiety disorders	17	30	4
Panic attacks	0	1	1
Stress disorders	3	6	11
Bipolar disorders	1	0	2
ADHD	1	0	2
Substance abuse	0	4	9

Screening tool
Using a current screening tool		Interested in using a new brief psychiatric tool
Yes	15(22.0%)	Yes	43(63.2%)
No	53(77.9%)	No	20(29.4%)
		Unsure	5(7.4%)

Psychiatric conditions treated

Depression and anxiety were the two most common conditions treated. For the PCPs who ranked the disorders, the majority (58.1%) ranked depression as the most common psychiatric condition treated, and 30.2% considered depression to be the second most common psychiatric condition treated. Approximately 35% ranked anxiety as the condition they most frequently treated, and the majority (62.7%) ranked anxiety as the second most common psychiatric condition treated. None of the PCPs ranked any of the following conditions as among the top three psychiatric conditions they treated: post‐traumatic stress disorder, obsessive–compulsive disorder, phobias or psychotic disorders (see Table 1 for the top three psychiatric conditions PCPs treat). The most common class of psychiatric medication prescribed was antidepressants. The percentage of PCPs who indicated that they prescribed a certain class of psychiatric medications was as follows: antidepressants (100%); anti‐anxiety (91.3%); sleep aids (87.9%); cognitive enhancers (77.5%); stimulants (43.1%); antipsychotics (41.3%) and mood stabilisers (36.2%; see Figure 1).

Figure 1.

Percentage of primary care physicians prescribing psychiatric medication

Practice habits

The psychiatric care practice habits of PCPs on average reflected the following: 40.1% of patients receive only medication; 29.6% receive medication and are also referred for therapy; 17.2% are referred to a psychiatrist; and 12.8% are referred just for therapy. As indicated by the data (see Table 1), prescribing psychiatric medication is the most typical method of treatment. Indeed, on average 70% of a primary care physician's mental health patients receive at least psychiatric medication as part of their psychiatric care. Overall, the least preferred treatment method is to refer patients for therapy alone. Interestingly, 13.7% of PCPs do not refer their patients to therapy and 6.8% do not prescribe psychiatric medications at all.

The psychiatric assessment practices of PCPs

The majority of physicians (77.5%) relied on traditional clinical interviewing to diagnose psychiatric problems. However, the majority (63.8%) indicated that they were willing to use a psychiatric screening tool (see Table 1). Although several psychiatric screening tools have been developed,¹⁴ PCPs are still relying heavily on traditional clinical interviewing to assess for psychiatric problems. The continued reliance of PCPs on traditional clinical interviewing may be due, in part, to a lack of efficient and user‐friendly psychiatric screening tools available in practice.

Discussion

Our findings showed that, similar to prior research, PCPs are treating approximately 30% of their patients for mental health problems,^21,28 and that depression and anxiety continue to be the most commonly treated conditions. Some research has suggested that the number of mental health patients treated in primary care may be as high as 70%²⁴ and that as many as 66–75% of all depression cases are treated by PCPs instead of by mental health providers.⁴ Our findings showed that PCPs are prescribing a wide range of psychiatric medications, and that for the majority of their mental health patients PCPs are serving as their patients' PPCPs. Additionally, PCPs tend to refer only a small percentage of their patients to mental health specialists, a finding consistent with prior research.³ Given our small sample size one should be cautious in generalising these findings. It is possible that a response bias existed in that those PCPs most invested in the psychiatric care of their patients decided to return the survey. Nonetheless, our findings concerning the number of psychiatric patients on an average PCP's caseload, the types of conditions treated and the types of psychiatric medications prescribed by PCPs are consistent with prior research.

The future of psychiatric care in primary care

Our aim in this study was to assess the overall psychiatric care practice habits of primary care physicians and to determine the extent to which there was a need for and interest in using a psychiatric screening tool. As indicated, the vast majority of PCPs are using standard clinical interviewing but would be interested in using a psychiatric assessment tool. However, despite the significant number of available screening tools, there does not appear to be a simple, effective and user‐friendly screening tool that can be efficiently incorporated into the busy practice of today's PCPs.¹⁴

Treating mental health patients in an already busy and stressful healthcare industry presents additional time management and treatment challenges for PCPs. To help improve the assessment of psychiatric conditions, physicians need a tool that is sufficiently uncomplicated and comfortable for their patients to complete, while still being easy and quick to interpret. Ultimately, the lack of a ‘usable’ psychiatric assessment tool to better assess patients' psychological health could result in improper use of resources and could contribute to patients' distress by prolonging the assessment process and delaying necessary treatment.^14,16,29,30

We believe that there is a strong need for PCPs to continue to function as the PPCP for those patients who will benefit from their PCP ‘skill mix’ and for those patients who have limited access to mental health care. The WHO/Wonca ²⁰ report of 2008 states that integrating mental health treatment into primary care is the most feasible way of closing the treatment gap and ensuring that people get the mental health care they need. However, to function successfully in the role of PPCP, when the average physical examination may last only 12–15 minutes,³ it would behove PCPs to enhance their psychiatric assessment methods through, for example, using a valid and efficient psychiatric screening tool. Having a proper psychiatric assessment tool that can be universally applicable, accurate and reliable is important in continuing to foster the successful integration of psychiatric care into primary care. To assist PCPs in detecting various psychiatric disorders and in initiating appropriate and timely treatment, we are developing a psychiatric assessment tool called the ‘ABCD Emotional Care Check‐Up’. The tool covers the eight most common psychiatric conditions seen in primary care in a user‐friendly and time‐efficient format and takes only three to five minutes to complete.

Ultimately, the successful integration of psychiatric care into primary care will be facilitated by developing better methods of psychiatric assessment. Incorporating a more standardised method of psychiatric assessment in primary care could assist PCPs in: 1) more successfully managing or preventing the patient's symptoms before they intensify into a more serious condition; 2) minimising out‐comes such as suicide or homicide by providing early intervention; 3) reducing self‐medicating, through for example, drug or alcohol abuse, by providing patients with alternative treatments such as psychiatric medications and therapy; 4) minimising liability by strengthening assessment of patients' psychiatric problems as well as strengthening documentation in the patients' records; and 5) improving treatment outcomes by collaborating more with mental health clinicians in coordinating more comprehensive psychiatric care for patients.

Appendix 1

Primary Care – Mental Health Patient Questionnaire

Age Range (circle one): 25–35 36–45 46–55 56–65 66–75 76+

Gender (circle one): Male Female Degree (circle one): MD DO

Specialty (circle one): PCP Other:______Years Practicing Medicine:__

1 What is the approximate # of patients on your caseload:______

2 Approximately what percentage of your patients present with a mental health problem (e.g. anxiety, depression, stress, sleep problems, psychosis, substance use, attention problems, anger problems)? ______%

3 Identifying mental health problems can occur in various ways. Do you use a mental health screening tool either at initial intake or for follow‐up visits to assist in identifying a mental health problem?

YES NO

If Yes, please name the screening tool:

______________________________________________

If No, please list the method you are using:

____________________________________________

4 Would you be interested in trying a new diagnostic screening tool to assist you in identifying and distinguishing among potential mental health problems in patients?

YES

NO

5 What problems are you treating in your mental health patients? Rank in order of most to least frequent (1 = most frequent problem, etc.). If you do not treat a problem below, simply leave it blank.

___Depression ___Anxiety ___Phobia ___Obsessive‐Compulsive Disorder

___PTSD ___ADHD ___Substance Abuse ___Schizophrenia

___Panic Attacks ___Bipolar ___Stress Condition ___Other

6 What classes of psychiatric medications do you prescribe? (circle all that apply):

Anti‐psychotics Anti‐depressants Anti‐anxieties

Stimulants Cognitive Enhancers (Alzheimer's)

Mood Stabilizers (includes Lithium) Sleep‐aids Others

7 Of your mental health patients, to what percentage do you: (distribution should equal 100%)

Only prescribe medication: ______%

Prescribe medication and refer for therapy: ______%

Refer just for therapy: ______%

Refer to a psychiatrist for treatment: ______%