Family Practice Advance Access originally published online on January 7, 2005
Family Practice 2005 22(1):37-42; doi:10.1093/fampra/cmh712
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Explaining variation in antidepressant prescribing rates in east London: a cross sectional study
Centre for General Practice and Primary Care, Institute of Community Health Sciences, Queen Mary's School of Medicine and Dentistry, University of London, Medical Sciences Building, Mile End Road, London E1 4NS, UK
Correspondence to Dr SA Hull; Email: s.a.hull{at}qmul.ac.uk
Received 15 August 2004; Accepted 27 September 2004.
Hull SA, Aquino P and Cotter S. Explaining variation in antidepressant prescribing rates in east London: a cross sectional study. Family Practice 2005; 22: 37–42.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionDeclarationReferences |
|---|
Background. Rates of depression and anxiety in south Asian populations are lower than expected. It remains uncertain whether this reflects a real difference in prevalence or differences in case recognition and management.
Objective. To examine whether concordance of culture or ethnicity between doctors and patients affects the prescribing rates for antidepressant and anxiolytic medications in general practice populations, taking into account demography, practice size and organization.
Method. A cross-sectional general practice study, using practice and demographic data from primary care trusts, doctors' place of qualification from the General Medical Council, combined with practice level prescribing data from the prescription pricing authority (PACT) for the period 2000–2002. Set in 139 practices in the east London primary care trusts (PCTs) of Tower Hamlets, Hackney and Newham, multiethnic areas with large populations of south Asian residents and doctors. The main outcome measure was the annual prescribing rates for each group of drugs, calculated as the mean of two years average daily quantities (ADQs) for each medication, divided by the practice population.
Results. In east London the median prescribing rate (ADQs) for all antidepressants was 7.97 (inter-quartile range 4.91–10.76), for all anxiolytics and hypnotics 2.27 (interquartile range 1.11–3.96). There were significant differences in prescribing rates between practices with UK trained GPs and practices with south Asian trained GPs, with the highest rates of antidepressant prescribing in practices with UK trained GPs and low proportions of south Asian patients. No differences were found in anxiolytic and hypnotic prescribing rates between these practices. 57% of the variation in prescribing between practices could be explained by a model including the place of GP qualification, the proportion of registered women, older (>65) patients, and the list size per full time GP.
Conclusions. Compared with previous studies prescribing rates for antidepressants have almost doubled over five years, the greatest increase being for selective serotonin re-uptake inhibitors (SSRIs). There is a modest fall in prescribing rates for anxiolytics and hypnotics. Concordance between south Asian practice populations and doctors from similar south Asian cultures is not associated with an increase in antidepressant prescribing. Lower rates of prescribing in practices with south Asian trained doctors occur regardless of the ethnic composition of the practice population. Reasons for these differences are uncertain, but may include differences in explanatory models for presenting symptoms, and management strategies which rely less on a biomedical paradigm.
Keywords. Antidepressant prescribing, Asian ethnicity, depression, general practice.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationReferences |
|---|
The recognition and management of depression in the general practice setting has been under scrutiny for some years. Initially the debate focussed on apparent low rates of recognition and the use of low doses and short courses of antidepressant medication, which ran counter to the evidence for the need for long term prescribing from studies based largely on hospital clinic populations.1 With increasing evidence on the effectiveness of psychological therapies, which many patients prefer, but which may be scarce and difficult to access, there is now debate whether GPs might be over-prescribing for mild and moderate depressive illness.2
Difficulties in identifying and managing depression may be greater in areas where there are large ethnic minority populations. These difficulties may include language and cultural barriers, differences in expectations of treatment and the time constraints of general practice.3 Debate continues on whether the apparent lower rates of depression and anxiety in south Asian groups is related to real differences in prevalence, to the difficulties in case recognition in cross cultural consultations, or to differences in explanatory models and the chosen source of help for mental health symptoms in different cultures.4 Existing studies on comparing prescribing rates between GPs of different ethnicity and place of qualification find little evidence of a difference in total prescribing rates, with the main determining factor for providing more prescriptions being higher levels of patient deprivation.5,6
Our previous research in east London found lower levels of antidepressant prescribing where practices had larger populations of south Asian patients.7 This study was developed to explore whether concordance between ethnicity of the doctor (using the indicators of place of birth and place of qualification) and practice population ethnicity was associated with higher rates of antidepressant prescribing.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationReferences |
|---|
Data on prescribing, practice characteristics and population factors for the registered practice populations were obtained for all 152 general practices in the east London boroughs of Newham, City and Hackney and Tower Hamlets. These are inner city localities with high levels of social deprivation, and an ethnically diverse population, the largest grouping being south Asian. Ethical approval was obtained from the north east London Health Authority Research Ethics Committee.
Practice and population data
The East London General Practice database project provided information on practice characteristics, including size, staffing and organisational factors.7,8 We used two measures of social deprivation attributed to practice populations from census data, the Townsend index as a measure of material deprivation and the low income scheme index (LISI), a measure related to our prescribing outcome variable.
Prescribing data
We obtained antidepressant and anxiolytics prescribing data from the Prescription Pricing Authority for the period July 2000 to June 2002 in the form of average daily quantities (ADQ). Aggregation of all the average daily quantities for a group of related drugs can provide a measure of the notional days of treatment for a therapeutic drug group in a practice population. The drug groups included all antidepressants, separated into tricyclics (TCAs), selective serotonin re-uptake inhibitors (SSRIs), monoamine oxidase inhibitors (MAOIs) and other antidepressants. Anxiolytics and hypnotics were combined. For each practice an annual prescribing rate for each of groups of drugs was calculated by dividing the total ADQs for each group by the practice population.
Population and doctor ethnicity
Details of practice population ethnicity from the 2001 census were not available to the project. As south Asian ethnicity was the focus of this study, we used an Asian naming programme to attribute the proportion of practice populations of Asian ethnicity. This method takes the names of patients on the practice list and assigns each one to an ethnic group. This method has been shown to be accurate for the attribution of south Asian ethnicity, but underestimates other groups such as Afro-Caribbean.9 We obtained data on place of birth of all principals in general practice from the east London information department (ELCHIS) and the place of primary medical qualification from the General Medical Council.
Statistical analysis
The unit of analysis was the general practice. Associations between practice and population characteristics and the main prescribing outcome variables were examined using descriptive statistics and simple linear regression models. More complex relationships, including sensitivity analyses, were examined using multiple regression. All statistical analysis was undertaken using Excel and Genstat.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationReferences |
|---|
Prescribing data were available for 143 practices in east London for the two year study period. Of these, four practices retired or dissolved during the study period, leaving 139 to be included in the analysis. Table 1 illustrates the range of practice prescribing for different groups of antidepressants and for all anxiolytics and hypnotics combined. The median ADQ of antidepressants per practice population was 7.97 (interquartile range 4.91–10.76). For anxiolytics the median ADQ was 2.27 (interquartile range 1.11–3.96).
|
Of the 352 general practice principals in these 139 practices, 35% had a south Asian place of qualification and 55% were qualified in the UK (see Table 2). There are potentially important differences between these groups of doctors. Whereas 35% of Asian qualified doctors work in single-handed practices (compared to 5% of UK qualified doctors), 60% of UK trained doctors work in training practices (compared to 6% of south Asian trained doctors). Table 2 shows the distribution of explanatory variables used in the analysis. A correlation matrix indicated associations between some of the explanatory variables; practices with a high proportion of south Asian qualified GPs are more likely to have higher proportions of south Asian patients (0.24), lower numbers of female patients (–0.21), and larger list sizes per full time equivalent GP (0.23). No associations were found between the rates of prescribing antidepressants and prescribing anxiolytics or hypnotics.
|
Figure 1 illustrates the relationship between the prescribing of antidepressants, the proportion of south Asian patients on practice lists, and the place of qualification of GPs. Where Asian populations are high, prescribing rates are low, but practices which have a low prescribing rate tend to have south Asian trained GPs regardless of the ethnicity of the practice population.
|
Multivariate analysis
The multivariate model (Table 4) using the explanatory variables in Table 3, was able to explain 57% of the variation between practice prescribing of antidepressant medication, with higher rates being associated with the proportion of GPs qualified in the UK, a higher proportion of the list being female, and larger numbers of older patients (>65 years). Smaller list sizes per GP were also associated with higher rates of prescribing. Anxiolytic and hypnotic prescribing was associated with the proportion of patients over 65 years of age, but this accounted for only 8% of the variation in prescribing.
|
|
The findings on antidepressant prescribing were explored with a further series of multivariate analyses. Using the data on GP place of birth explained less of the variance than using place of qualification. In order to reduce the confounding effect between the size of the south Asian practice population and south Asian GPs we explored a model limited to the practices which had 20% or less south Asian patients; this had no effect on the explanation of variance. A model based only on the adult population (over 16 years) also gave similar results.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationReferences |
|---|
Prescribing trends
Our study hypothesis, that where there was evidence of south Asian cultural concordance between practice populations and GPs antidepressant prescribing might be higher, was not borne out. Our findings are more complex. Among practices where the doctors gained their primary medical qualification in south Asia, prescribing rates for antidepressants, but not for anxiolytics or hypnotics, are lower than for doctors trained in the UK. This difference is independent of the tendency of south Asian patients to register at practices with Asian doctors.
Previous studies have shown higher rates of prescribing in general for Asian populations compared to white populations, which is likely to reflect both higher morbidity and high consultation rates.10,11,12 Studies on antidepressant prescribing have identified lower rates where Asian populations are high.7 Our results illustrate the importance of including characteristics of the doctor alongside those of the patient and the illness.
Antidepressant prescribing in east London has almost doubled since 1996, with median rates rising from 4.13 in 1996 to 7.97 in 2001.7 This rise is similar to national trends, with prescriptions for selective serotonin reuptake inhibitors rising fastest.2 Using an estimate of 5% of the population with a major depressive disorder, this is the equivalent of 160 antidepressant prescribing days for each individual. The median prescribing rate for anxiolytics and hypnotic has fallen slightly in the same period of time. We have found no evidence of an association between antidepressant and anxiolytics prescribing, so our results do not suggest that GPs who prescribe low rates of antidepressants substitute with anxiolytics.
Study limitations
The attribution of ethnicity and its use in observational studies is fraught with problems both of theory and method. Our aim was to capture shared aspects of culture which might contribute to the ease of identifying depressive illness, which may in turn be associated with the prescription of antidepressant medication. We considered that both place of birth and place of medical education of the study GPs would be the most likely variables to affect prescribing patterns. However south Asia is a large geographic area and we have not been able to take account of the cultural diversity within the region. In spite of these reservations we find that south Asian populations are preferentially registered with south Asian doctors. There are systematic differences in the practice characteristics (particularly size, training status, and numbers of female patients) of the UK and south Asian practices. Some of the differences in antidepressant prescribing may be related to these factors, but none of these differences were reflected in anxiolytic and hypnotic prescribing rates.
We did not systematically capture data on other ethnic groups in practice populations. However practice localities with known higher proportions of black residents did not show any associations with prescribing rates. We did not include details of counsellors or clinical psychologists linked to practices, but previous work suggest this does not affect prescribing.13
Understanding the context of prescribing
GPs form part of the narrative chain for patients with common mental disorders. This extends from defining and naming symptoms of illness, through negotiating and conceptualizing a domain of causation (which in current western medicine may be biological, psychological or social) and the proper or accepted modality of response or treatment. All of these processes are subject to cultural influences by both doctors and patients.14,15 Two examples may be relevant to our findings. Explanatory models of illness vary between cultures.16 Exploration of the explanatory models used to understand experiences of mental distress by different ethnic groups in east London suggests that Afro-Caribbean and Bangladeshi groups gave significantly more spiritual and supernatural explanations for their symptoms than White groups. Bangladeshis also believed that effective treatments focused on the spiritual dimension and on physical treatments for the body (Ruddel K, personal communication). In a detailed study of Punjabi general practice attendees Bhui found that Punjabi doctors did not recognise more depression than other Asian doctors.17 In common with other studies, the distress of south Asian patients was construed more in physical than psychiatric terms compared to the white attendees. This may be a reflection of the doctors' own cultural beliefs which preferentially construe distress without resort to illness labeling or the use of medication.
We know little about the ways that different cultural groups self manage symptoms of ‘minor mental illness’, but it may be that techniques which ameliorate symptoms within family networks or a spiritual framework are as effective as a biological or psychological explanatory and management framework. The call for a rethink on the increasing rates of antidepressant prescribing for mild and moderate depression may be timely in view of their of depression.
|
Declaration |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationReferences |
|---|
Funding: PA was funded by Newham primary care trust. SC is supported by the North East London Consortium for Research and Development.
Ethical approval: obtained from the north east London Health Authority Research Ethics Committee.
Conflicts of interest: none.
|
Acknowledgments |
|---|
We are grateful to the late Paul Datlen, who collated much of the data for the study. Also to Clive Ball, information manager for ELCHIS, and the prescribing advisors at the three study PCTs. SH and PA designed the study, PA and Paul Datlen collected the data. SC provided statistical analysis. SH and PA wrote the paper with contributions from SC.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationReferences |
|---|
1 Paykel ES, Priest RG. Recognition and management of depression in general practice: consensus statement. Br Med J 1992; 305: 1198–1202.
2 Mild depression in general practice: time for a rethink? Drugs and Therapeutic Bulletin 2003; 4: 60–64.
3 Commanders MJ, Sashi Dharan SP, Odell SM, Surtes SG. Access to mental health care in an inner city health district: pathways into and within specialist psychiatric services. Br J Psychiatry 1997; 170: 312–316.
4 Nazroo JY. Ethnicity and mental health: findings from a national community survey. Policy Studies Institute; 1999.
5 Gill P, Dowell A, Harris C. Effects of doctors' ethnicity and country of qualification on prescribing patterns in single handed general practices: linkage of information collected by questionnaire and from routine data. Br Med J 1997; 315: 1590–1594.
6 Gill P, Dowell A, Harris C. The effect of doctor ethnicity and country of qualification on prescribing patterns: an ecological study. J Clin Pharm and Ther 1999; 24: 197–199.
7 Hull SA, Cornwell J, Harvey C, Eldridge S. Prescribing rates for psychotropic medication among east London general practices: low rates where Asian populations are greatest. Fam Pract 2001; 18: 167–172.
8 Hull SA, Jones IR, Moser K. Factors influencing the attendance rates at accident and emergency departments in east London: the contributions of practice organisation, population characteristics and distance. J Health Serv Res Policy 1997; 2: 6–14.[Medline]
9 Nicoll A, Bassett K, Ulijaszek SJ. What's in a name? J Epidemiol Com Health 1986; 40: 364–368.
10 Naish J, Sturdy P, Bobby J, Pereira P. The association between Asian ethnicity and prescribing rates in east London general practices: a database study. Health Informatics J 1998; 100–105.
11 Gillam SJ, Jarman B, White P, Law R. Ethnic differences in consultation rates in urban general practice. Br Med J 1989; 299: 953–957.
12 Balarajan R. Ethnicity and variations in the nation's health. Health Trends 1995; 4: 114–119.
13 Sibbald B, Addington–Hall J, Brenneman D, Freeling P. Investigation of whether on site general practice counsellors have an impact on psychotropic drug prescribing rates and costs. Br J Gen Pract 1996; 46: 63–67.[Web of Science][Medline]
14 Kleinman A. Anthropology and psychiatry, the role of culture in cross-cultural research on illness. Br J Psychiatry 1987; 151: 447–454.
15 Bose R. Psychiatry and the popular conception of possession among the Bangladeshis in London. Int J Soc Psychiatry 1997; 43: 1–15.
16 Jacob KS, Bhugra D, Lloyd KR, Mann AH. Common mental disorders, explanatory models and consultation behaviour among Indian women living in the UK. J Royal Soc Med 1998; 91: 66–71.[Abstract]
17 Bhui K, Bhugra D, Goldberg D, Dunn G, Desai M. Cultural influences on the prevalence of common mental disorder, general practitioners' assessments and help-seeking among Punjabi and English people visiting their general practitioner. Psychological Med 2001; 31: 815–825.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. Kumarapeli, R. Stepaniuk, S. de Lusignan, R. Williams, and G. Rowlands Ethnicity recording in general practice computer systems J Public Health, September 1, 2006; 28(3): 283 - 287. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||