Knowledge of the Edinburgh Postnatal Depression Scale among psychiatrists and general practitioners

The aim was to examine the impression that both general practitioners and psychiatrists have a low level of awareness and knowledge of the Edinburgh Postnatal Depression Scale (EPDS), have little experience in its use, and would not feel confident in giving advice on issuesarising from its use by health visitors. Postnatal depression (PND) is a common condi tion affecting mothers following 10-15% of all deliveries (Kumar & Robson, 1984) and is poorly recognised for several reasons. Healthcare pro fessionals are found to concentrate on physical illness and the baby, and so neglect maternal mental health (Goldberg, 1992). Mothers feel guilty about discussing their ambivalent and negative feelings with partners and health professionals. Its recognition is important since the associated morbidity not only adversely affects the mother (Watson et al, 1984) but also the developing child (Murray & Stein. 1989). It is important to promote awareness and increase detection of PND. which is readily treatable. self-report scoring highly (> 12/13) on the EPDS which correlate with PND.

Postnatal depression (PND) is a common condi tion affecting mothers following 10-15% of all deliveries (Kumar & Robson, 1984) and is poorly recognised for several reasons. Healthcare pro fessionals are found to concentrate on physical illness and the baby, and so neglect maternal mental health (Goldberg, 1992). Mothers feel guilty about discussing their ambivalent and negative feelings with partners and health professionals. Its recognition is important since the associated morbidity not only adversely affects the mother (Watson et al, 1984) but also the developing child (Murray & Stein. 1989). It is important to promote awareness and increase detection of PND. which is readily treatable.
The Edinburgh Postnatal Depression Scale (EPDS) is a research screening instrument developed by Cox and co-workers (1987) in Edinburgh and consists of a ten-item selfreport questionnaire.
It has high specificity and sensitivity for identifying possible cases of postnatal depression, which are then diag nosed by clinical interview. It has been validated as a research screening instrument for use within the UK population (Cox et al 1987) and within most Western cultures, although cross-cultural research outwith wes tern Europe and North America remains in its infancy (O'Hara. 1994). It has been used and validated in clinical practice in the UK primary care setting where it should be administered by a health professional experienced in its use, most often a health visitor. In clinical use there is a need for clearly identified referral systems (Holden, 1994) to manage cases scoring highly (> 12/13) on the EPDS which correlate with PND. As part of their continuing professional devel opment, health visitors in our local health district have been instructed in the use of the EPDS and taught non-directive counselling skills (Holden. 1989) with the aim of improving the recognition, assessment and management of PND. They have adopted the EPDS for general clinical use. Their expectation is that general practitioners (GPs) and psychiatrists in our area will be familiar with the EPDS and its use and be able to give advice on issues arising.
The GPs and psychiatrists from whom they might expect advice and support are anecdotally reported to be unfamiliar with the instrument or the rationale for its use. We surveyed GP and psychiatrist knowledge of the EPDS.

The study
A letter/questionnaire was sent to 173 psychia trists at all grades in the St George's Hospital training rotation, including 39 consultant psy chiatrists based within St George's Hospital and Pathfinder Mental Health NHS Trust, 48 senior registrars, and the remainder registrars, senior house officers and non-consultant grades who either worked within St George's Hospital or Pathfinder Mental Health NHS Trust, or who would rotate there as part of training and may be subject to the St George's academic pro gramme, their clinical commitments being within Merton, Sutton and Wandsworth. London. We also wrote to 350 of the 353 principals in general practice listed with the Merton, Sutton and Wandsworth Health Authority.
The letter/questionnaire described the reason for the inquiry and asked the respondent to answer five questions. These were: (1) I am aware of the Edinburgh Postnatal Depression Scale (Yes/No).
(2) I have used the Edinburgh Postnatal Depression Scale (Yes/No). (3) A score of 14 represents health (Yes/No/ Don't know). (4) I would feel confident now to give advice on issues arising from the use of the Edin burgh Postnatal Depression Scale (Yes/ No). (5) What issue does question 10 address? (or indicate don't know).
The reply letter was anonymous except that respondents could be identified as GPs or psychiatrists.
All were provided with an ad dressed envelope in which to return the com pleted questionnaire. Respondents were invited to identify themselves on a separate piece of paper only if they wished to receive further information on the EPDS. Table 1 shows the response rate to the survey, and Table 2 the survey results. Only 57 (46%) psychiatrists and 47 (33%) GP respondents were aware of the EPDS. In both groups, fewer than 6% of respondents had used the scale or were happy to advise on issues arising from its use. Less than 6% could identify the nature of question 10, the suicide/self-harm item. Margin ally more (6%) identified a total score of 14 as not indicating health.

Comment
The letter/questionnaire was restricted to one side of A4 and styled to require the best response to be ringed for the first four questions, with a written response only required for the fifth. Response rates to the questionnaire letter were good for psychiatrists (71%) and poorer for GPs (41%), the latter in keeping with the limitations of the postal questionnaire for surveys (Bennett & Ritchie. 1975).
The difference in response rates between psychiatrists and GPs could be explained by non-enclosure of a stamp on the self-addressed envelope to GPs. The majority of psychiatrists replied by internal post.
The psychiatrist group contained trainees who may have more time and inclination to answer research questionnaires. Assessing the number of consultant replies was only possible for those who identified their status when requesting further information -7 out of the 39.
GPs and psychiatrists vary in their attitudes to mental illness, reflecting their different experi ence in practice (Kerr et al, 1995). GPs are found to be more inclined to audit care of physical illness than psychiatic illness, possibly because of the lack of objective measures of psychiatric disorder available for use in general practice (Wright & Anderson. 1995). Psychiatrists should be familiar with the use of psychiatric assess ment instruments.

Conclusions
This study indicates a remarkably low level of knowledge, skill and familiarity with the EPDS among GPs and psychiatrists in the Merton, Sutton and Wandsworth health authority area. It is unlikely that either professional group, with out further training, could support health visitors in the use of the EPDS.
There is an urgent need to embark on training for GPs and community mental health team (CMHT) psychiatrists in the use of the EPDS, and to provide specialist advice when requested. This approach would have the attraction of including a wider body of healthcare profes sionals in the recognition and management of PND, using and enhancing skills already pos sessed, and reducing the number of specialist referrals. The cost involved in providing the educational opportunity to GPs and psychia trists should be considered in relation to the cost of health visitors using a scale which, without opportunity for further treatment, is of no benefit to the mother.
Expansion of tertiary mother and baby psy chiatry services could include direct routes of communication and referral for health visitors using the EPDS. This has the advantage of making specialist knowledge directly available to the primary healthcare worker, but has two disadvantages. First, it may bypass and poten tially de-skill GPs and CMHT psychiatrists. Second, it may generate inappropriate referrals that could be assessed and managed adequately  (37%)57 (46%)5(4%)4(3%)6(5%)3(2%)3(2%)59(41%)47 (33%)8(6%)1 1 (8%)6(5%)6(5%)6(5%) in the primary care setting, without overburden ing the specialist service. The initiative from health visitors is welcome. Experience and knowledge of the EPDS should advance through its wider clinical use and inclusion in local education programmes. Health visitors must exercise extreme caution in using a screening instrument in an area where support from other agencies is uninformed or unavail able. This can be overcome by close liaison with specialist mother and baby mental health ser vices in joint initiatives to inform and support GPs and community mental health teams.