Printer Friendly

Increasing the Continuity of Care between Primary Care Provider and a Psychiatric Hospital in Singapore.


Continuity of care is vital to the mental health of service users with complex needs. (1-4) While system integration poses an important challenge for continuity of care, (5) in Singapore, the issue of continuity of care poses a greater challenge because of the relative nascence of its community mental health services. Furthermore, traditional Chinese beliefs may impact help seeking behaviour in populations with mental illness, especially older individuals. (6) As a result of the development of institutionalised care in Singapore, hospitals remain central to the daily treatment of mental illness. In other healthcare delivery contexts, these illnesses could be treated by a general practitioner (GP). As the awareness of mental illness increases and the wide treatment gap closes, (7) the demand for services will increase. (8) It will therefore become important to share the task of treating mental illness. Sharing this task will require the proliferation of partnerships between specialists within hospitals and primary care providers in the community.

In Singapore, the current model of care distributes the responsibility of providing healthcare services among various service providers. Our institute provides specialist outpatient and tertiary care for acute psychiatric illnesses and crisis interventions. Two satellite clinics offer outpatient care to geographically distant locations. These services are subsidised by government-provided means-tested financial assistance. (9,10) The level of subsidy varies across other service providers, and receiving subsidy in one clinic or institute does not guarantee receiving the subsidy elsewhere. Public clinics, known as polyclinics, provide general medical care and some specialised services, which vary across clinics, but usually include paediatric care, dental care, imaging and laboratory facilities, and nursing care. Specialist psychiatric services are usually absent. Historically, polyclinics and individual GPs situated within the community provide primary care for a minority of persons with chronic mental illness. However, some service providers lack the confidence in providing the care necessary to maintain clinical stability and wellness, a finding echoed in other contexts. (11)

This current method of distributing responsibility and delivering care does not fully take advantage of the professional capabilities of the clinicians working in the community-integrated polyclinics. For stable mental health outpatients, maintaining a follow-up with a primary care provider or family physician is sufficient to meet their psychiatric treatment needs. Furthermore, physicians working within the polyclinics are equipped with the core medical knowledge necessary to meet these needs.

To close the gap between community and institutional services, our institute created a GP-partnership programme to link service users with primary care providers. The primary care providers receive training at our institute to update their psychiatric knowledge. They receive continued support from our institute, in the form of an open channel of communication, available whenever necessary (usually by phone or email). Our institute refers stable service users to the GP-partnership programme to facilitate their access to community health services and to reduce the burden on hospital services. Given the high rate of metabolic co-morbidities, (12) service users are encouraged to take advantage of the GP-partnership as the physicians may be better positioned to treat metabolic co-morbidities. People are eligible for GP-partnership referral if they meet the following criteria: service users must be stable, require only maintenance dose medications, and have not been hospitalised within the past 6 months. They are ineligible if they have disruptive personality disorders, have a history of suicide attempts or aggression, or if they are receiving clozapine or benzodiazepines (both of which have specific monitoring protocols within our institute).

The challenge arises during the transition from the institutional follow-up to the primary care provider follow-up. Despite the partnership programme, the rate at which service users actually follow through with the referrals remains unsatisfactory. Improving these rates is the target of our quality improvement project reported here.


We derived our data from administrative records, which documented the number of referrals accepted by service users. Our dataset extended from April 2014 to June 2016. Data were extracted in aggregate, with no individual-level data used in the analysis. Our outcome of interest was successful referral to the GP, defined as the service user attending the service provider to which they were referred on the date of their appointment. As the project constituted a quality improvement project, and the case managers providing the intervention had ethically permissible access to the population, we did not seek ethics approval from the local review boards. Under Singapore standards and as per the guidelines issued by the National Healthcare Group Domain Specific Review Board, such quality improvement projects are exempt from ethics review. Instead, the Institute's Clinical Practice Improvement Programme (CPIP) review committee granted project approval. The quality improvement project complied with the stipulations of the CPIP review committee. This study was done in accordance with the principles outlined in the Declaration of Helsinki.

We surveyed 25 service users referred to the GP-partnership programme from December 2014 to January 2015 to explore their willingness to follow through with the referral made. We targeted equal numbers of people who were successfully referred and people who did not follow through with their referral. Fifteen service users were willing to follow through with the referral and joined the programme. They noted the importance of convenient location, out-of-hours service, and most importantly to them, the importance of avoiding the stigma associated with receiving follow-up care at a mental health institute. This was an unexpected motivating factor for use of the GP-partnership programme. Ten service users declined GP follow-up, noting that they believed GPs to be expensive (due to lack of subsidies) and inexperienced with mental illness. The service user responses resonated with the experiences of team members. By means of the Pareto chart, group voting and the Pareto 80-20 heuristic (13) we identified the causes of referral failures and isolated 3 potential areas of improvement: (1) concerns about the cost of GP services; (2) uncertainty about financial subsidies; and (3) concerns about GP's inexperience in mental health care.

Interventions to Improve Uptake of Referrals

Based on the abovementioned potential areas of improvement, the following 3 interventions were planned and implemented in March, June, and September 2015.

First, to address misconceptions and concerns about the increased cost of seeing a GP rather than a physician at our institute, we implemented tailored financial counselling. Case managers identified suitable service users, checked the charges service users would have to pay at the GP (adjusted for the type of financial subsidies service users would receive), and explained to them the components of the charges. Tailoring the explanation of costs was essential, as standard published charges may be general and result in misleading estimations of costs. This tailored financial counselling intervention proactively gave the information service users needed to correct their misconceptions about the cost of seeing a GP. Such service users were not likely to search for the necessary information on their own, and it was beneficial to have a trusted professional (case manager) take the time to explain the charges.

The second intervention was built upon our first intervention and sought to raise awareness of various financial schemes that could help service users cover the costs of GP services. This intervention provided education and in-vivo assistance with key processes. For the service users who had declined the GP referral because of costs, case managers or executives informed them of the potential alternative forms of subsidisation. For some of them, such information was sufficient. However, others required greater assistance and the case manager completed the necessary subsidy application forms along with the service users. The application process, notorious for its complexity, acted as a barrier for those with the greatest need. Service users who had been given assistance were contacted 2 weeks after our visit to determine if they had successfully applied for the subsidies. The support provided to service users during the application process has become standard for all case managers working with this population. This intervention was then widely implemented to all service users referred to the programme, including those who previously rejected the programme.

The third intervention sought to correct misconceptions about primary service provider's skills and capacity to treat mental illness. It was important to note to service users that the primary care providers to which they were referred had undergone specific training with our institute, and that they continued to receive support from our institute to maintain their clinical skills. The channels the institute had established with the primary care providers reassured service users of the quality of care they would receive from these service providers. Furthermore, we established avenues for 24-hour assistance should the primary care providers require it. This newly established programme 'hotline' was also the means by which service users could rejoin our services without obtaining a new referral, eliminating concerns about being excluded from our services. Finally, for those who were reluctant to follow up with their referral to primary care providers because of concerns about GP qualifications, case managers arranged for regular contact between the service user and the case manager, and between the primary care provider and the staff managing the GP-partnership programme. This contact was structured to have a minimum of one contact prior to the first appointment, one contact immediately after the first appointment, one contact after 3 visits to the GP, and one contact every 6 months to maintain a link with the hospital. The 6-month contact is perpetual and changes only if the individual returns to the hospital.


We used a single-group interrupted time-series analysis (14) to determine the impact of the intervention period and post-intervention period on the trend observed prior to our project. Interrupted time-series analysis allows us to accommodate correlation between time points. We chose to segment the timeline into pre-intervention period (April 2014 to February 2015) and post-intervention period (March 2015 to June 2016). The intervention period grouped all 3 interventions to allow sufficient observations to be included in the segment. We lagged the model by 1 month to allow the emergence of effects from the intervention. Regression coefficients were estimated by ordinary least-squares regression and the Newey-West method was used for deriving the standard errors. As a balance measure, we looked at the proportion of people who returned to the hospital following inadequate primary care provider management or a deterioration in mental health. Analyses were conducted in STATA 13, with the ITSA package SJ16-3 st0389_2.


During 11 months of pre-intervention period (April 2014 to February 2015), 238 people representing a mean ([+ or -] standard deviation) of 22 [+ or -] 4 new cases per month (range, 16-27 cases; interquartile range [IQR], 19-25 cases) were potentially suitable for referral to GP services. Of these, 152 people, representing a mean of 14 [+ or -] 2 new cases per month (range, 11-18 cases; IQR, 12-16 cases) successfully followed up with referrals made by psychiatrists to obtain follow-up care from the GP programme, which accounted for a 64% success rate. Our interventions aimed to increase this figure to 100%. In all, 26% (135/521) of our service users were male and their mean age was 54 [+ or -] 15.8 years. The distribution of mental health conditions resembled that of our general service users (15) in terms of the frequency of schizophrenia, depression and bipolar disorder, with the exception of anxiety which was present in 17% (91/521) of cases (compared with 3% in general service user); 14% (72/521) had diagnoses of other disorders. Figure 1 presents the proportion of service users who followed through with the referral and Figure 2 illustrates the interrupted time series over the observation period. Specific regression values are given in the Table. Over the pre-intervention period the rate of increase in accepted referrals was non-significant (0.29% per month, 95% confidence interval: -0.20 to 0.78%; p = 0.23), suggesting the system had a stable follow-up rate prior to our quality improvement project. Following the intervention, the figure increased by 12% (95% confidence interval: 5.3-18.9%; p = 0.001). The monthly rate of increase over the intervention period was significant (1.4% per month, 95% confidence interval: 0.7-2.1%; p < 0.001).

Over the 16 months following the implementation of the first intervention, we were able to increase the proportion of service users who had followed up their referral to a GP to 92% (n = 260 / 283), a monthly mean of 16 out of 17 referrals. Increasing the rate of successful referrals did not reduce the percentage of people who had to return to the hospital following a deterioration of their mental health. During baseline, 89% of service users referred to the GP programme remained with the GP over the follow-up period. Following our intervention, this proportion remained at 90%. We therefore found no change in our balance measure.

We accomplished our high rate of referral acceptance by addressing issues of low confidence in primary care provider skill and by providing counselling on the various subsidies for which our service users were eligible. This improvement was statistically and clinically significant. We believe that all 3 interventions contributed to the success of the project, and that implementing any one of the these interventions without considering the challenges addressed by the others would likely yield lower rates of successful referrals.


When used in combination, our strategies achieve a higher rate of referral follow-through to GP services by people with mental health disorders. Addressing concerns about the cost of GP services by providing financial counselling helped people understand their entitlement to social assistance programmes. However, since financial assistance programmes change over time, this requires continued efforts to maintain current knowledge about government programmes. Addressing concerns about GP expertise in the management of mental health disorders satisfied reticent service users and improved their willingness to receive services from primary care providers. Leveraging what consumers actually value has been shown to be more effective at producing change than leveraging what physicians may think their consumers value. (16) The entire case management department has adopted the strategies applied in our quality improvement project, and these strategies have since remained in effect.

It is important to note that many people successfully followed through with referrals because GP services were geographically easier to access. For those who were employed, the extended office hours of the GP programme also provided convenient appointments. This follows standard elements of continuity of care. (1,2) However, the stigma of attending services offered at a mental health hospital also motivated service users to obtain treatment elsewhere. (17) Case managers must be mindful of the stigma that surrounds attendance at mental health services because it may lead to a reticence to seeking psychiatric care (7) in the event of deterioration in the person's mental health.

The barriers to successful adherence with referrals highlighted by our project echo independent findings in the same service user population. (16) Research following a rational patient model (18) attempting to explain adherence to treatments highlighted that cost was a barrier to care, but also highlighted that variables which fall outside the rationale patient model contribute to service user's desire to adhere with prescribed courses of action. Specifically, the emotions surrounding stigma and people's beliefs of the qualifications of service providers may sway their decision to follow up with a new referral. Addressing this latter point allowed us to successfully increase the number of completed referrals, highlighting the importance of addressing emotional components of service user's decision making.

We found it important to regularly update case managers' and programme executives' knowledge of the financial schemes that could help our service users. Given that eligibility rules and allowances changed yearly, it may be possible that certain people become eligible for financial assistance without their knowledge. Updating the documents and pamphlets distributed to service users was also important to reduce financial barriers to services. While such a step may appear obvious, case managers and programme executives need to add this item to their long list of items with which they need to stay current. Justifying the added effort may be difficult if service providers do not take ownership of the role (19) or if there is belief that financial considerations fall outside their area of responsibility.

Reassurance has always been an important aspect of case management, (20,21) and the findings of our quality improvement project indicate that service seekers need reassurance about their service eligibility in addition to reassurance about their mental health status. Reassuring service users about the qualifications and the training of the GPs to which they were being referred remained a constant necessity. In addition, reassuring service users about their continued eligibility for our institutional services served to reduce their reluctance to transfer to new, unfamiliar service providers. These efforts to reassure service users point to the importance of engaging with their needs and responding accordingly. Maintaining a good mechanism for feedback is an important way in which service providers can stay informed about the changing needs of service users.


We tried to sample equal numbers of people who followed through with their referral to primary care services, and of those who did not. However, obtaining feedback from those who did not follow through with the referral was difficult. It is possible that this minor imbalance skewed the issues we leveraged to effect change. This is unlikely given that the obstacles we addressed have been highlighted elsewhere in research conducted on similar populations. (16) To preserve the anonymity of the sample, we did not extract individual-level data from the administrative records. We were therefore unable to determine if individual-level variables influenced the probability of successfully following up with referrals.

Conclusion and Implications

Increasing the number of people seen for routine follow-up care by primary care providers increases the capacity of the mental healthcare system. Given that financial support entitlements change, it is important for case managers to remain aware of changing policy to be able to facilitate continuity of care between institutional and community care. Misconceptions of service provider qualifications may have a great impact on service user's willingness to seek help from primary care providers. If case managers address these misconceptions, follow-up rates for referrals to primary care providers may improve. These efforts are also likely to reduce the treatment gap associated with having multiple conditions, by providing low-barrier access alternatives that are also stigma-free, especially when paired with community capacity. (22)

Strengthening connections with GPs was essential to ensuring referrals were accepted, and ensuring the care service users received when they had successfully been referred remained adequate to maintain their mental health status. Increasing service provider exposure to people with mental illness will likely contribute to a de-stigmatising effect, as professional and personal exposure has been shown to reduce stigma among service providers. (23) These connections also served to widen the pool of GPs open to receiving referrals from the institute. By word of mouth within the community of GPs, strong ties served to spread awareness of the training and support available from our institute to reduce any concerns about lack of professional knowledge and experience related to the treatment of people with mental illness.

Transferring the responsibility of providing maintenance care to people with stable mental illness to primary care providers increases the capacity of the mental healthcare system as a whole. It is expected that the quality of care is adequate to sustain the mental health of service users although that remains to be demonstrated in our locality in future studies.


We would like to thank Drs Alvin Lum and Christine Tan for their support with the Clinical Practice Improvement Programme. We would also like to thank Drs Alex Su and Bernard Wong for their advice and encouragement. Lastly, we would like to thank Joshua Wee and Gina Teo for helping us provide the financial counselling interventions to the study participants.


Dr Daniel Poremski was partially funded by the Singapore Ministry of Health's National Medical Research Council under the Centre Grant Programme (Grant No.: NMRC/CG/004/2013). The Granting organisation played no role in the execution analysis or reporting of the presented data.


(1.) Adair CE, McDougall GM, Mitton CR, Joyce AS, Wild TC, Gordon A, et al. Continuity of care and health outcomes among persons with severe mental illness. Psychiatr Serv 2005;56:1061-9.

(2.) Bachrach LL. Continuity of care for chronic mental patients: a conceptual analysis. Am J Psychiatry 1981;138:1449-56.

(3.) Dewa CS, Jacobson N, Durbin J, Lin E, Zipursky RB, Goering P. Examining the effects of enhanced funding for specialized community mental health programs on continuity of care. Can J Commun Ment Health 2010;29(S5):23-40.

(4.) Puntis S, Rugkasa J, Forrest A, Mitchell A, Burns T. Associations between continuity of care and patient outcomes in mental health care: a systematic review. Psychiatr Serv 2015;66:354-63.

(5.) Durbin J, Goering P, Streiner DL, Pink G. Does systems integration affect continuity of mental health care? Adm Policy Ment Health 2006;33:705-17.

(6.) Found A. Relationship between traditional Chinese beliefs about aetiology of mental disorders and help seeking: a survey of the elderly in Macao. East Asian Arch Psychiatry 2016;26:3-9.

(7.) Chong SA, Abdin E, Sherbourne C, Vaingankar J, Heng D, Yap M, et al. Treatment gap in common mental disorders: the Singapore perspective. Epidemiol Psychiatr Sci 2012;21:195-202.

(8.) Rose N. Historical changes in mental health practice. In: Thornicroft G, Szmukler GI, Mueser KT, Drake RE, editors. Oxford textbook of community mental health. Oxford: Oxford University Press; 2011: 9-18.

(9.) Medifund. Ministry of Health Singapore. Available from: Accessed 4 Nov 2015.

(10.) McKee M, Busse R. Medical savings accounts: Singapore's non-solution to healthcare costs. BMJ 2013;347:f4797.

(11.) Carr VJ, Lewin TJ, Barnard RE, Walton JM, Allen JL, Constable PM, et al. Attitudes and roles of general practitioners in the treatment of schizophrenia compared with community mental health staff and patients. Soc Psychiatry Psychiatr Epidemiol 2004;39:78-84.

(12.) Yogaratnam J, Biswas N, Vadivel R, Jacob R. Metabolic complications of schizophrenia and antipsychotic medications--an updated review. East Asian Arch Psychiatry 2013;23:21-8.

(13.) World Health Organization. Quality improvement in primary health care: a practical guide. World Health Organization, Regional Office for the Eastern Mediterranean; 2004.

(14.) Linden A. Conducting interrupted time series analysis for single and multiple group comparisons. Stata J 2015;15:480-500.

(15.) Poremski D, Kunjithapatham G, Koh D, Lim XY, Alexander M, Lee C. Lost keys: understanding service providers' impressions of frequent visitors to psychiatric emergency services in Singapore. Psychiatr Serv 2017;68:390-5.

(16.) Poremski D, Sagayadevan VD, Wang P, Lum A, Subramaniam M, Chong SA. The impact of stakeholder preferences on service user adherence to treatments for schizophrenia and metabolic comorbidities. PLoS One 2016;11:e0166171.

(17.) Chong SA, Abdin E, Vaingankar JA, Kwok KW, Subramaniam M. Where do people with mental disorders in Singapore go to for help? Ann Acad Med Singapore 2012;41:154-60.

(18.) Corrigan PW, Rusch N, Ben-Zeev D, Sher T. The rational patient and beyond: implications for treatment adherence in people with psychiatric disabilities. Rehabil Psychol 2014;59:85-98.

(19.) Isobel S, Edwards C. Using trauma informed care as a nursing model of care in an acute inpatient mental health unit: a practice development process. Int J Ment Health Nurs 2017;26:88-94.

(20.) Rapp CA. The active ingredients of effective case management: a research synthesis. Community Ment Health J 1998;34:363-80.

(21.) Intagliata J, Baker F. Factors affecting case management services for the chronically mentally ill. Adm Ment Health 1983;11:75-91.

(22.) Cheang KM, Cheok CC. A 17-month review of the care model, service structure, and design of THRIVE, a community mental health initiative in Northern Singapore. East Asian Arch Psychiatry 2015;25:168-74.

(23.) Mittal D, Ounpraseuth ST, Reaves C, Chekuri L, Han X, Corrigan P, et al. Providers' personal and professional contact with persons with mental illness: relationship to clinical expectations. Psychiatr Serv 2016;67:55-61.

H Huang, D Poremski, YL Goh, M Hendriks, D Fung

Ms Hannah Huang, BSc, Institute of Mental Health, Singapore.

Dr Daniel Poremski, PhD, Institute of Mental Health, Singapore.

Dr Yen-Li Goh, MBBS, MMed (Psychiatry), Institute of Mental Health, Singapore.

Ms Margaret Hendriks, RN, RMN, BHS (Nursing), Institute of Mental Health, Singapore.

Dr Daniel Fung, MMed (Psychiatry), Institute of Mental Health, Singapore.

Address for correspondence: Dr Daniel Poremski, 10 Buangkok View, Singapore 539747. Tel: (65) 63893634; Email:

Submitted: 20 February 2017; Accepted: 20 October 2017
Table. Regression coefficients for interrupted time-series analysis.

                             Coefficient (%)  Newey-West      T Value
                                              standard error

Intercept                    62.31            1.66            37.58
Slope of first time series    0.29            0.24             1.23
Change at the interruption   12.09            3.30             3.66
Slope of second time series   1.40            0.33             4.24

                             p Value  95% Confidence
                                      interval (%)

Intercept                    < 0.001  58.88 to 65.74
Slope of first time series     0.23   -0.20 to 0.78
Change at the interruption     0.001   5.26 to 18.92
Slope of second time series  < 0.001   0.72 to 2.08
COPYRIGHT 2017 The Hong Kong College of Psychiatrists Ltd.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2017 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Special Communication
Author:Huang, H.; Poremski, D.; Goh, Y.L.; Hendriks, M.; Fung, D.
Publication:East Asian Archives of Psychiatry
Article Type:Report
Geographic Code:9SING
Date:Dec 1, 2017
Previous Article:Fear of Fear and Broad Dimensions of Psychopathology over the Course of Cognitive Behavioural Therapy for Panic Disorder with Agoraphobia in Japan.
Next Article:Feasibility and Clinical Utility of High-definition Transcranial Direct Current Stimulation in the Treatment of Persistent Hallucinations in...

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters