A Critical Analyses of a midwfery aticle
...by some practitioners it must be accurate and a complete summary of the full article (Polit &Hungler, 1991). Again it can be said that the article achieves this model. It emphasises the research issue and the sample population, it mentions the methodology approach with reference to the survey, the results and conclusions. On the other hand the accuracy of the abstract is questioned in view of the percentage rates stated in the results of the findings. Four of the percentage rates discussed in the abstract does not correspond with the percentage rates documented in the main article. A second issue is that the abstract does not refer to the fact that the findings need to be interpreted with caution as there are two important limitations of the article. Introduction to the research article The introduction sets the scene for the article (Drummond, 1996). It should discuss the theoretical background of the problem under consideration and evaluate research done previously (Polgar &Thomas, 1991). The article clearly discusses the background to the study and the rationale why the study has been ventured upon, making references to related research. It recognises the important, professional issues and the key concepts relevant to the profession or clinical practice (Rees, 2004). The aims of the research which are to describe : the role and experience of midwives in identifying and referring women with possible depression in pregnancy and after birth and to also describe current policies and practices in maternity units for the care of women with depression follow logically from the original problem. The original problem being that there is little empirical evidence to support the effectiveness of universal screening for depression either antenatally or postnatally and it is questionable how much education midwives possess or to what extent they are actually involved in the screening process. It is a requirement of the midwives role to keep knowledge and skills up-to-date throughout their working life, in particular to take part regularly in learning activities that develop competence and performance (NMC, 2002). There is no obvious review made of previous research literature concerned with topics of the same nature. In view of this it is thought that the researcher avoids reading too much literature prior to data collection in case it influences the way data are collected and prejudges what are deemed to be important issues. The researcher will allow the participants to define what is important (Rees, 2004). Nevertheless, a poor literature review does not invalidate the findings whether the reason for the poor literature review is the researchers disregard or ignorance of other work, or a desire not to be influenced by others (Parahoo & Reid, 1988). Methodology The Criteria set out to show the methodology behind the research needs to demonstrate the type of research, the sample size, who was included and why, research tools used and the way the data was collected and analysed (Ogier, 1989). The above criteria have been accurately discussed in the article making it easy for any other researcher to conduct the same study at another time in order to make comparisons of the findings (Cluett &Bluff, 2000). A qualitative research method was used in the form of a grounded theory approach. The aim being to produce an explanation or theory of the role of midwives in depression screening in the findings through the researchers interpretation and analysis (Rees, 2003). Qualitative research involves broadly stated questions about human experiences and realities, studied through sustained contact with people in their natural environments, generating rich, descriptive data that help us to understand their experiences (LoBiondo-Woods & Haber, 1994). This is demonstrated in the article by quotes and comments made by midwives that were collected via a questionnaire. The aim of this research tool is a characteristic of a qualitative approach and is used to find out peoples feelings and experiences from their own point of view rather than that of the researcher, however interviews provide more flexibility and depth, and are far more in keeping with the philosophy of the approach (Rees C, 1996). It can be suggested that the results analysed from questionnaires can be subjective in a way of the researcher being selective in the sampling, therefore by combining both questionnaires with interviews the study could have been strengthened. The tool of data collection used in the survey was post questionnaires to all maternity units in England and Wales. The questionnaires were addressed to the head of midwifery or senior midwife in each unit. The covering letter requested that the questionnaires were completed by the midwife with the most experience of postnatal depression (PND) services. This method possesses a number of problems within the information gathered for this article and for replicating the study at a later date. There is no guarantee that the most suitable person who completed the questionnaire did so in isolation or that it was given to the most appropriate person to complete. Some of the maternity units may have felt the need to promote their unit and to give the best impression, therefore there may have been a collaboration of heads that completed the questionnaires which could result in giving answers that do not demonstrate a completely true reflection. In other words this could be a biased response. The researchers sent out reminders after two weeks to units that had not responded to the questionnaires which in turn achieved an 86% response rate. There is no clear definition of what is an acceptable response rate but it is suggested that a response rate of 75%-85% from a postal survey is very good (Treece &Treece, 1986). However, there is no reference made to the non-respondents within the statistical information which suggests that a totally different result may have been given to the content of the article. It is proposed that it is important to know as much as possible about non-responders, as those who do not respond may differ from those who do and so introduce bias in the sampling (Moser & Kalton, 1971). Respondents could have been compared to non-respondents to establish similarity, so reducing the risk of bias. The final problem with the use of the questionnaires is the fact that the article pays no reference as to where to find the questions that were asked and therefore the quality of the questionnaire can not be analysed to see if it is written in a way to interest and motivate the respondents to reply positively. It is suggested that a questionnaire should not normally take longer than 20-25 minutes to complete (Treece & Treece, 1986). It is also suggested that questions should not be any longer than 20 words (Oppenheim, 1992). The use of quotes gives ‘life’ to the article and should provide support for the interpretation of the data (Cluett & Bluff, 2000). The quotes give the reader the impression that midwives are very much aware that there is a need for education and support systems within a multidisciplinary team with regards to referring cases of women with depression. It is perceived that this would enable midwives to carry out the role of screening for depression to a higher level. However, due to the anonymity of the source, the quotes may be questioned as there may be other reasons why the negative quotes are provided, for example there could be poor working relationship with the midwife and the units hierarchy. The sample size is composed of midwives employed in 211 maternity units in England and Wales. The size of the sample could be considered too small to be able to draw firm conclusions due to the fact that the sampling frame appears to be biased. The reason for this is the fact that the researchers selected only the midwives with the relevant knowledge of PND and guidelines for referral. This promotes some doubt as to whether the sample is appropriate which could adversely influence the validity and reliability of the study. It is necessary when planning the study that the researchers consider the characteristics that make individuals eligible for selection, and those that would make them unsuitable or even put them at risk or at a disadvantage (Rees, 2003). It is thought that the researchers would have an indication of the results therefore by excluding midwives lacking knowledge of the screening process and the inability to respond adequately this enhanced the results and provided justification for carrying out the study. There is no mention that a pilot study was carried out in order to check the consistency of the tool of data collection. This is probably mainly due to the fact that the qualitative researcher does not develop a standard tool of data collection derived from the literature review, or use a tool that has been validated in previous research. This enables the researcher to keep an open mind in important issues within the study, therefore it is not possible to carry out a pilot study as there is no structured research tool to test for reliability (Rees, 2003). Ethical considerations It is not clear whether ethical committee approval was gained. The ethical considerations apply to all research that involves the participant of clients and approval from the appropriate committee must be sought. Confidentiality and anonymity must be maintained and clear and concise information must be given to all participants to enable them the freedom from exploitation (Cluett & Bluff, 2000). As there appears to be no ethical approval for the study this can cause dilemmas, particularly as the sample was small. It would be unethical to undertake research on clients if the sample was too small to provide useful information (Cluett, 1996). Results The research results show that an 86% response rate was achieved from the questionnaires used to format the article and identifies the different response rates in regions from the lowest to the highest. No reason for further research has been suggested as to why there were such variations which ranged from 66% - 100%. The article provides information on 5 main domains being antenatal and postnatal depression screening, referral, information available to women, training and problems encountered by midwives. The article findings are shown in table form showing both percentages and the raw data so that the actual size of the denominator can be determined as this varies throughout the table due to respondents not completing all the questions asked of them. The results however do not appear to consider the amount of drop outs in the tests used and who undertakes the screening for antenatal depression as this falls from 182 respondents to 47 who answered questions. The results proceed to discuss what appears to show a distinct lack of specific training for midwives but as training is sometimes only felt as training when in a lecture type environment rather than on the job further consideration needs to be given to these results. Articles discussion The discussion begins by referring to regional differences between policies and practices in the maternity units around the country. These differences are highlighted as cultural, personnel or the availability of resources within the different trusts. It was found that although 95% of units questioned women at the booking session about previous or current psychological difficulties, only 25% of units undertook formal screening for all women. It was found interesting that health visitors undertook more screening than midwives although health visitors have less contact with women during the antenatal period than midwives. Again, lack of time, training and information was proposed as the reason why. It was a surprising factor to find that only two- thirds of the units had leaflets available for antenatal depression and just over half had leaflets available for postnatal depression. The discussion acknowledges limitations of the study and recommends a need for future research as it is beyond the scope of the survey to distinguish the barriers as to why the responding midwives gave the reasons for the difficulties they experience with depression screening and caring for women in their care. Conclusions In view of the readability of the article it can be said that it flows well and is easy to read, however the results sections are confusing, containing a large amount of statistics and percentages. The research was well designed but the researchers can be seen to have been too ambitious as there seems to be an overload of data resulting in a lack of depth. Well-designed research that did not go according to plan is not wasted if it can be replicated in another study (Tarling & Crofts, 2002). The findings reliably show that the policies for screening need to be reviewed as it is evident that midwives need time given to training packages in order to demonstrate competencies within this area. However, it can be argued whether all midwives need training or should specialists be assigned to this role, for example a mental health nurse. By using this policy it may strengthen the effectiveness of the referral systems and as the study suggests that without good collaboration of referral policies screening is liable to cause more harm than good. If this is the case, it can be said that research is required into the effectiveness of routine screening in order to produce evidence-based practice rather than replicating studies of this nature. This article was critiqued in comparison with ‘Midwives and perinatal mental health’ (Stewart & Henshaw, 2002) a local study, and ‘Malaysian midwives’ views on postnatal depression (Keng, 2005) a national, up-to-date study. Both discuss similar findings which reinforce the above conclusions. References Cluett E, (1996), Introduction to statistics 1, Modern Midwife, 6 (9) p30-34 Cluett E R, Bluff R (2000), Critiquing the literature, Principles and Practice of Research in Midwifery, Bailliere Tindill Department of Health (1993), The Report of the Expert Maternity Group, Changing Childbirth, London, HMSO Department of Health (2004), Why Mothers Die 2000-2002, 6th Report of the Confidential Enquiries into Maternal Deaths in the United Kingdom, Royal College of Obstetricians and Gynaecologists, Press, London Drummond A (1996), Research Methods for Therapists, Cheltenham, Stanley Thomas Ltd, (R850) Gyte G (1994), Putting Research into Practice in Maternity Care, Modern Midwife, 4 (8) p19-20 Keng S L (2005), Malaysian Midwives’ views on Postnatal Depression, British Journal of Midwifery, 13 (2) p78-85 Lobiondo-Woods G, Haber J (1994), Methods, Critical Appraisal and Utilisation, Nursing Research, 3rd edn, Mosby, St. Louis Moser C A, Kalton G (1971), Survey Methods in Social Investigation, 2nd edn, Heinmann, London Nursing and Midwifery Council (NMC), (2002), Code of Professional Conduct, London Oglier M (1989), Reading Research, Scutari Press, London Oppenheim A N, (1992), Questionnaire Design, Interviewing and Attitude Measurement, 2nd edn, London Parahoo K, Reid N (1988), Research Skills, Nursing Times, 84 (43) 69-71 Polgar S (2000), Introduction to Research in the Health Sciences, Churchill Livingstone Price A, Price B (1996), Critical Reading, Modern Midwife, 7 (5) p26-29 Rees C (1996), Quantitative and Qualitative Approaches to Research, British Journal of Midwifery, 4(7) p374-377 Rees C (1997), Introduction to Research for Midwives, Books for Midwives, London Rees C (2003), Critiquing Research Articles, Introduction to Research for Midwives, Books for Midwives, Edinburgh Siddiqui J (1994) A Philosophical Exploration of Midwifery Knowledge, British Journal of Midwifery, 2 (9) p149 Stewart C, Henshaw C (2002), Midwives and Perinatal Mental Health, British Journal of Midwifery, 10 (2) p117-121 Tarling M, Crofts L (2002), Reviewing the Literature, The Essential Researchers Handbook for Nurses and Health Care Professionals, 2nd edn, Bailliere Tindill Treece E W, Treece JW (1986), Elements of Research in Nursing, C V Mosby, St, Louis Walton I, Hamilton M (1995), Midwives and Changing Childbirth, Books for Midwives, London Tully L, Garcia J, Davidson L, Marchant S (2002), Role of Midwives in Depression Screening, British Journal of Midwifery, 10 (6) p374-378 CRITIQUE OF A RESEARCH ARTICLE There is evidence that perinatal mental health disorders are present at significant rates in the population. Regardless of this there are no structured evidence-based practices in place to ensure women who suffer receive appropriate care. Maternal suicides have been reported to the Confidential Enquiries into Maternal Deaths over the last fifty years, however it is only in the last three enquiries that the cause of psychiatric deaths have been separately looked at and that a psychiatrist has been a member of the enquiry along with a central assessor. The mental health and wellbeing of women in pregnancy is central to secure good clinical, social outcomes for the mother and baby (DoH, 2004). It is therefore important to identify any problems from mental health at the earliest stage possible. This requires a structured screening process throughout maternity units to ensure that women who are suffering or previously suffered from a mental illness are given the appropriate care. It has been recommended that midwives should distinguish between postnatal depression, severe depressive illness, puerperal psychosis and other psychiatric disorders and follow up with a General Practitioner or psychiatric services (DoH, 2004). The midwives role is to promote the interests of patients and clients, this includes helping individuals and groups gain access to health and social care, information and support relevant to their needs (NMC, 2002) It is seen that more research is required into the midwives role of depression screening and that midwives should be able to critically read research articles (Walton & Hamilton, 1995, Price, 1996). Although research articles are important it must be remembered that research evidence cannot be substituted for decision making (Rees, 1997). Women who access the maternity services may be guided by research findings, however, they will make decisions based on their own beliefs, instincts, wishes and priorities (Gyte, 1994). There is a crucial requirement to recognise the body of knowledge that supports the practice of midwifery. It is therefore essential that midwifery embraces an epistemology of practice which is based upon reflection and relevant research. It is important that research is ventured upon for the best reasons to improve care for mothers and babies (Siddiqui, 1994). It is of vital significance to know whether a particular practice is effective, to what extent there may be any side effects and to appreciate women’s thoughts about the type of care they receive. Practice should be based on sound evidence (DoH, 1993). This essay will critique an article published in the British Journal of Midwifery titled ‘Role of midwives in depression screening’ (Tully et al, 2002) It is intended to go further than understanding the article so as to examine if from different viewpoints in order to provide a critique of findings and propose possible errors and false presumptions. It looks to analyse the way the article has been written, highlighting ways in which the article excels as well as fails in terms of readability and accuracy. It is the aim of the writer to ensure that this critique will be constructive, penetrating and decisive and to appreciate its relevance to midwifery practice and education (Cluett &Bluff, 2000). This will lead to a conclusion as to the overall significance and acceptance of the article. Title The title of the article is concise and informative. It is however a misleading as implies that the article will discuss the role of the midwife when screening women for depression. In fact the article is based on a survey to determine which current policies and practices are adopted at maternity units throughout the country to identify antenatal or postnatal depression. Therefore, although the title is concise it does lack clarity and fails to use a precise use of words to convey the meaning intended (Cluett &Bluff, 2000). Although it is ambiguous, and goes some way to reflect the articles content, this does not mean that the work done is necessarily poor (Drummond 1996). The title makes no mention of the research approach but it may be perceived that due to the nature of the title a more qualitative approach would be expected. Authors The authors contributing to writing the article consist of a researcher from the National Perinatal Epidemiology Unit (NPEU), from Oxford, a social scientist from the NPEU, Oxford, a Director from the NPEU, Oxford and a midwife researcher from the NPEU, Oxford. Due to the fact that the National Perinatal Epidemiology Unit is a well respected unit that specialises in maternity care research it can be said that their skills may potentially heighten the quality of the study (Cluett & Bluff, 2000). However, as the article presents findings analysed from questionnaires in the form of a frequency table, it can also be argued that the quality of the study may be have been further strengthened if there had been a statistician involved. Even so, this does not mean that any of the other researchers lack the skills to analyse the data. The article was published in the British Journal of Midwifery, 2002 which is an appropriate journal for the literature to be published in. Abstract An abstract should be short, concise and not exceed 250 words and should provide the reader with a description of the entire report (Polgar & Thomas, 1991), this abstract fulfils this criteria. As the abstract is the only part of article read by some practitioners it must be accurate and a complete summary of the full article (Polit &Hungler, 1991). Again it can be said that the article achieves this model. It emphasises the research issue and the sample population, it mentions the methodology approach with reference to the survey, the results and conclusions. On the other hand the accuracy of the abstract is questioned in view of the percentage rates stated in the results of the findings. Four of the percentage rates discussed in the abstract does not correspond with the percentage rates documented in the main article. A second issue is that the abstract does not refer to the fact that the findings need to be interpreted with caution as there are two important limitations of the article. Introduction to the research article The introduction sets the scene for the article (Drummond, 1996). It should discuss the theoretical background of the problem under consideration and evaluate research done previously (Polgar &Thomas, 1991). The article clearly discusses the background to the study and the rationale why the study has been ventured upon, making references to related research. It recognises the important, professional issues and the key concepts relevant to the profession or clinical practice (Rees, 2004). The aims of the research which are to describe : the role and experience of midwives in identifying and referring women with possible depression in pregnancy and after birth and to also describe current policies and practices in maternity units for the care of women with depression follow logically from the original problem. The original problem being that there is little empirical evidence to support the effectiveness of universal screening for depression either antenatally or postnatally and it is questionable how much education midwives possess or to what extent they are actually involved in the screening process. It is a requirement of the midwives role to keep knowledge and skills up-to-date throughout their working life, in particular to take part regularly in learning activities that develop competence and performance (NMC, 2002). There is no obvious review made of previous research literature concerned with topics of the same nature. In view of this it is thought that the researcher avoids reading too much literature prior to data collection in case it influences the way data are collected and prejudges what are deemed to be important issues. The researcher will allow the participants to define what is important (Rees, 2004). Nevertheless, a poor literature review does not invalidate the findings whether the reason for the poor literature review is the researchers disregard or ignorance of other work, or a desire not to be influenced by others (Parahoo & Reid, 1988). Methodology The Criteria set out to show the methodology behind the research needs to demonstrate the type of research, the sample size, who was included and why, research tools used and the way the data was collected and analysed (Ogier, 1989). The above criteria have been accurately discussed in the article making it easy for any other researcher to conduct the same study at another time in order to make comparisons of the findings (Cluett &Bluff, 2000). A qualitative research method was used in the form of a grounded theory approach. The aim being to produce an explanation or theory of the role of midwives in depression screening in the findings through the researchers interpretation and analysis (Rees, 2003). Qualitative research involves broadly stated questions about human experiences and realities, studied through sustained contact with people in their natural environments, generating rich, descriptive data that help us to understand their experiences (LoBiondo-Woods & Haber, 1994). This is demonstrated in the article by quotes and comments made by midwives that were collected via a questionnaire. The aim of this research tool is a characteristic of a qualitative approach and is used to find out peoples feelings and experiences from their own point of view rather than that of the researcher, however interviews provide more flexibility and depth, and are far more in keeping with the philosophy of the approach (Rees C, 1996). It can be suggested that the results analysed from questionnaires can be subjective in a way of the researcher being selective in the sampling, therefore by combining both questionnaires with interviews the study could have been strengthened. The tool of data collection used in the survey was post questionnaires to all maternity units in England and Wales. The questionnaires were addressed to the head of midwifery or senior midwife in each unit. The covering letter requested that the questionnaires were completed by the midwife with the most experience of postnatal depression (PND) services. This method possesses a number of problems within the information gathered for this article and for replicating the study at a later date. There is no guarantee that the most suitable person who completed the questionnaire did so in isolation or that it was given to the most appropriate person to complete. Some of the maternity units may have felt the need to promote their unit and to give the best impression, therefore there may have been a collaboration of heads that completed the questionnaires which could result in giving answers that do not demonstrate a completely true reflection. In other words this could be a biased response. The researchers sent out reminders after two weeks to units that had not responded to the questionnaires which in turn achieved an 86% response rate. There is no clear definition of what is an acceptable response rate but it is suggested that a response rate of 75%-85% from a postal survey is very good (Treece &Treece, 1986). However, there is no reference made to the non-respondents within the statistical information which suggests that a totally different result may have been given to the content of the article. It is proposed that it is important to know as much as possible about non-responders, as those who do not respond may differ from those who do and so introduce bias in the sampling (Moser & Kalton, 1971). Respondents could have been compared to non-respondents to establish similarity, so reducing the risk of bias. The final problem with the use of the questionnaires is the fact that the article pays no reference as to where to find the questions that were asked and therefore the quality of the questionnaire can not be analysed to see if it is written in a way to interest and motivate the respondents to reply positively. It is suggested that a questionnaire should not normally take longer than 20-25 minutes to complete (Treece & Treece, 1986). It is also suggested that questions should not be any longer than 20 words (Oppenheim, 1992). The use of quotes gives ‘life’ to the article and should provide support for the interpretation of the data (Cluett & Bluff, 2000). The quotes give the reader the impression that midwives are very much aware that there is a need for education and support systems within a multidisciplinary team with regards to referring cases of women with depression. It is perceived that this would enable midwives to carry out the role of screening for depression to a higher level. However, due to the anonymity of the source, the quotes may be questioned as there may be other reasons why the negative quotes are provided, for example there could be poor working relationship with the midwife and the units hierarchy. The sample size is composed of midwives employed in 211 maternity units in England and Wales. The size of the sample could be considered too small to be able to draw firm conclusions due to the fact that the sampling frame appears to be biased. The reason for this is the fact that the researchers selected only the midwives with the relevant knowledge of PND and guidelines for referral. This promotes some doubt as to whether the sample is appropriate which could adversely influence the validity and reliability of the study. It is necessary when planning the study that the researchers consider the characteristics that make individuals eligible for selection, and those that would make them unsuitable or even put them at risk or at a disadvantage (Rees, 2003). It is thought that the researchers would have an indication of the results therefore by excluding midwives lacking knowledge of the screening process and the inability to respond adequately this enhanced the results and provided justification for carrying out the study. There is no mention that a pilot study was carried out in order to check the consistency of the tool of data collection. This is probably mainly due to the fact that the qualitative researcher does not develop a standard tool of data collection derived from the literature review, or use a tool that has been validated in previous research. This enables the researcher to keep an open mind in important issues within the study, therefore it is not possible to carry out a pilot study as there is no structured research tool to test for reliability (Rees, 2003). Ethical considerations It is not clear whether ethical committee approval was gained. The ethical considerations apply to all research that involves the participant of clients and approval from the appropriate committee must be sought. Confidentiality and anonymity must be maintained and clear and concise information must be given to all participants to enable them the freedom from exploitation (Cluett & Bluff, 2000). As there appears to be no ethical approval for the study this can cause dilemmas, particularly as the sample was small. It would be unethical to undertake research on clients if the sample was too small to provide useful information (Cluett, 1996). Results The research results show that an 86% response rate was achieved from the questionnaires used to format the article and identifies the different response rates in regions from the lowest to the highest. No reason for further research has been suggested as to why there were such variations which ranged from 66% - 100%. The article provides information on 5 main domains being antenatal and postnatal depression screening, referral, information available to women, training and problems encountered by midwives. The article findings are shown in table form showing both percentages and the raw data so that the actual size of the denominator can be determined as this varies throughout the table due to respondents not completing all the questions asked of them. The results however do not appear to consider the amount of drop outs in the tests used and who undertakes the screening for antenatal depression as this falls from 182 respondents to 47 who answered questions. The results proceed to discuss what appears to show a distinct lack of specific training for midwives but as training is sometimes only felt as training when in a lecture type environment rather than on the job further consideration needs to be given to these results. Articles discussion The discussion begins by referring to regional differences between policies and practices in the maternity units around the country. These differences are highlighted as cultural, personnel or the availability of resources within the different trusts. It was found that although 95% of units questioned women at the booking session about previous or current psychological difficulties, only 25% of units undertook formal screening for all women. It was found interesting that health visitors undertook more screening than midwives although health visitors have less contact with women during the antenatal period than midwives. Again, lack of time, training and information was proposed as the reason why. It was a surprising factor to find that only two- thirds of the units had leaflets available for antenatal depression and just over half had leaflets available for postnatal depression. The discussion acknowledges limitations of the study and recommends a need for future research as it is beyond the scope of the survey to distinguish the barriers as to why the responding midwives gave the reasons for the difficulties they experience with depression screening and caring for women in their care. Conclusions In view of the readability of the article it can be said that it flows well and is easy to read, however the results sections are confusing, containing a large amount of statistics and percentages. The research was well designed but the researchers can be seen to have been too ambitious as there seems to be an overload of data resulting in a lack of depth. Well-designed research that did not go according to plan is not wasted if it can be replicated in another study (Tarling & Crofts, 2002). The findings reliably show that the policies for screening need to be reviewed as it is evident that midwives need time given to training packages in order to demonstrate competencies within this area. However, it can be argued whether all midwives need training or should specialists be assigned to this role, for example a mental health nurse. By using this policy it may strengthen the effectiveness of the referral systems and as the study suggests that without good collaboration of referral policies screening is liable to cause more harm than good. If this is the case, it can be said that research is required into the effectiveness of routine screening in order to produce evidence-based practice rather than replicating studies of this nature. This article was critiqued in comparison with ‘Midwives and perinatal mental health’ (Stewart & Henshaw, 2002) a local study, and ‘Malaysian midwives’ views on postnatal depression (Keng, 2005) a national, up-to-date study. Both discuss similar findings which reinforce the above conclusions. References Cluett E, (1996), Introduction to statistics 1, Modern Midwife, 6 (9) p30-34 Cluett E R, Bluff R (2000), Critiquing the literature, Principles and Practice of Research in Midwifery, Bailliere Tindill Department of Health (1993), The Report of the Expert Maternity Group, Changing Childbirth, London, HMSO Department of Health (2004), Why Mothers Die 2000-2002, 6th Report of the Confidential Enquiries into Maternal Deaths in the United Kingdom, Royal College of Obstetricians and Gynaecologists, Press, London Drummond A (1996), Research Methods for Therapists, Cheltenham, Stanley Thomas Ltd, (R850) Gyte G (1994), Putting Research into Practice in Maternity Care, Modern Midwife, 4 (8) p19-20 Keng S L (2005), Malaysian Midwives’ views on Postnatal Depression, British Journal of Midwifery, 13 (2) p78-85 Lobiondo-Woods G, Haber J (1994), Methods, Critical Appraisal and Utilisation, Nursing Research, 3rd edn, Mosby, St. Louis Moser C A, Kalton G (1971), Survey Methods in Social Investigation, 2nd edn, Heinmann, London Nursing and Midwifery Council (NMC), (2002), Code of Professional Conduct, London Oglier M (1989), Reading Research, Scutari Press, London Oppenheim A N, (1992), Questionnaire Design, Interviewing and Attitude Measurement, 2nd edn, London Parahoo K, Reid N (1988), Research Skills, Nursing Times, 84 (43) 69-71 Polgar S (2000), Introduction to Research in the Health Sciences, Churchill Livingstone Price A, Price B (1996), Critical Reading, Modern Midwife, 7 (5) p26-29 Rees C (1996), Quantitative and Qualitative Approaches to Research, British Journal of Midwifery, 4(7) p374-377 Rees C (1997), Introduction to Research for Midwives, Books for Midwives, London Rees C (2003), Critiquing Research Articles, Introduction to Research for Midwives, Books for Midwives, Edinburgh Siddiqui J (1994) A Philosophical Exploration of Midwifery Knowledge, British Journal of Midwifery, 2 (9) p149 Stewart C, Henshaw C (2002), Midwives and Perinatal Mental Health, British Journal of Midwifery, 10 (2) p117-121 Tarling M, Crofts L (2002), Reviewing the Literature, The Essential Researchers Handbook for Nurses and Health Care Professionals, 2nd edn, Bailliere Tindill Treece E W, Treece JW (1986), Elements of Research in Nursing, C V Mosby, St, Louis Walton I, Hamilton M (1995), Midwives and Changing Childbirth, Books for Midwives, London Tully L, Garcia J, Davidson L, Marchant S (2002), Role of Midwives in Depression Screening, British Journal of Midwifery, 10 (6) p374-378 CRITIQUE OF A RESEARCH ARTICLE There is evidence that perinatal mental health disorders are present at significant rates in the population. Regardless of this there are no structured evidence-based practices in place to ensure women who suffer receive appropriate care. Maternal suicides have been reported to the Confidential Enquiries into Maternal Deaths over the last fifty years, however it is only in the last three enquiries that the cause of psychiatric deaths have been separately looked at and that a psychiatrist has been a member of the enquiry along with a central assessor. The mental health and wellbeing of women in pregnancy is central to secure good clinical, social outcomes for the mother and baby (DoH, 2004). It is therefore important to identify any problems from mental health at the earliest stage possible. This requires a structured screening process throughout maternity units to ensure that women who are suffering or previously suffered from a mental illness are given the appropriate care. It has been recommended that midwives should distinguish between postnatal depression, severe depressive illness, puerperal psychosis and other psychiatric disorders and follow up with a General Practitioner or psychiatric services (DoH, 2004). The midwives role is to promote the interests of patients and clients, this includes helping individuals and groups gain access to health and social care, information and support relevant to their needs (NMC, 2002) It is seen that more research is required into the midwives role of depression screening and that midwives should be able to critically read research articles (Walton & Hamilton, 1995, Price, 1996). Although research articles are important it must be remembered that research evidence cannot be substituted for decision making (Rees, 1997). Women who access the maternity services may be guided by research findings, however, they will make decisions based on their own beliefs, instincts, wishes and priorities (Gyte, 1994). There is a crucial requirement to recognise the body of knowledge that supports the practice of midwifery. It is therefore essential that midwifery embraces an epistemology of practice which is based upon reflection and relevant research. It is important that research is ventured upon for the best reasons to improve care for mothers and babies (Siddiqui, 1994). It is of vital significance to know whether a particular practice is effective, to what extent there may be any side effects and to appreciate women’s thoughts about the type of care they receive. Practice should be based on sound evidence (DoH, 1993). This essay will critique an article published in the British Journal of Midwifery titled ‘Role of midwives in depression screening’ (Tully et al, 2002) It is intended to go further than understanding the article so as to examine if from different viewpoints in order to provide a critique of findings and propose possible errors and false presumptions. It looks to analyse the way the article has been written, highlighting ways in which the article excels as well as fails in terms of readability and accuracy. It is the aim of the writer to ensure that this critique will be constructive, penetrating and decisive and to appreciate its relevance to midwifery practice and education (Cluett &Bluff, 2000). This will lead to a conclusion as to the overall significance and acceptance of the article. Title The title of the article is concise and informative. It is however a misleading as implies that the article will discuss the role of the midwife when screening women for depression. In fact the article is based on a survey to determine which current policies and practices are adopted at maternity units throughout the country to identify antenatal or postnatal depression. Therefore, although the title is concise it does lack clarity and fails to use a precise use of words to convey the meaning intended (Cluett &Bluff, 2000). Although it is ambiguous, and goes some way to reflect the articles content, this does not mean that the work done is necessarily poor (Drummond 1996). The title makes no mention of the research approach but it may be perceived that due to the nature of the title a more qualitative approach would be expected. Authors The authors contributing to writing the article consist of a researcher from the National Perinatal Epidemiology Unit (NPEU), from Oxford, a social scientist from the NPEU, Oxford, a Director from the NPEU, Oxford and a midwife researcher from the NPEU, Oxford. Due to the fact that the National Perinatal Epidemiology Unit is a well respected unit that specialises in maternity care research it can be said that their skills may potentially heighten the quality of the study (Cluett & Bluff, 2000). However, as the article presents findings analysed from questionnaires in the form of a frequency table, it can also be argued that the quality of the study may be have been further strengthened if there had been a statistician involved. Even so, this does not mean that any of the other researchers lack the skills to analyse the data. The article was published in the British Journal of Midwifery, 2002 which is an appropriate journal for the literature to be published in. Abstract An abstract should be short, concise and not exceed 250 words and should provide the reader with a description of the entire report (Polgar & Thomas, 1991), this abstract fulfils this criteria. As the abstract is the only part of article read by some practitioners it must be accurate and a complete summary of the full article (Polit &Hungler, 1991). Again it can be said that the article achieves this model. It emphasises the research issue and the sample population, it mentions the methodology approach with reference to the survey, the results and conclusions. On the other hand the accuracy of the abstract is questioned in view of the percentage rates stated in the results of the findings. Four of the percentage rates discussed in the abstract does not correspond with the percentage rates documented in the main article. A second issue is that the abstract does not refer to the fact that the findings need to be interpreted with caution as there are two important limitations of the article. Introduction to the research article The introduction sets the scene for the article (Drummond, 1996). It should discuss the theoretical background of the problem under consideration and evaluate research done previously (Polgar &Thomas, 1991). The article clearly discusses the background to the study and the rationale why the study has been ventured upon, making references to related research. It recognises the important, professional issues and the key concepts relevant to the profession or clinical practice (Rees, 2004). The aims of the research which are to describe : the role and experience of midwives in identifying and referring women with possible depression in pregnancy and after birth and to also describe current policies and practices in maternity units for the care of women with depression follow logically from the original problem. The original problem being that there is little empirical evidence to support the effectiveness of universal screening for depression either antenatally or postnatally and it is questionable how much education midwives possess or to what extent they are actually involved in the screening process. It is a requirement of the midwives role to keep knowledge and skills up-to-date throughout their working life, in particular to take part regularly in learning activities that develop competence and performance (NMC, 2002). There is no obvious review made of previous research literature concerned with topics of the same nature. In view of this it is thought that the researcher avoids reading too much literature prior to data collection in case it influences the way data are collected and prejudges what are deemed to be important issues. The researcher will allow the participants to define what is important (Rees, 2004). Nevertheless, a poor literature review does not invalidate the findings whether the reason for the poor literature review is the researchers disregard or ignorance of other work, or a desire not to be influenced by others (Parahoo & Reid, 1988). Methodology The Criteria set out to show the methodology behind the research needs to demonstrate the type of research, the sample size, who was included and why, research tools used and the way the data was collected and analysed (Ogier, 1989). The above criteria have been accurately discussed in the article making it easy for any other researcher to conduct the same study at another time in order to make comparisons of the findings (Cluett &Bluff, 2000). A qualitative research method was used in the form of a grounded theory approach. The aim being to produce an explanation or theory of the role of midwives in depression screening in the findings through the researchers interpretation and analysis (Rees, 2003). Qualitative research involves broadly stated questions about human experiences and realities, studied through sustained contact with people in their natural environments, generating rich, descriptive data that help us to understand their experiences (LoBiondo-Woods & Haber, 1994). This is demonstrated in the article by quotes and comments made by midwives that were collected via a questionnaire. The aim of this research tool is a characteristic of a qualitative approach and is used to find out peoples feelings and experiences from their own point of view rather than that of the researcher, however interviews provide more flexibility and depth, and are far more in keeping with the philosophy of the approach (Rees C, 1996). It can be suggested that the results analysed from questionnaires can be subjective in a way of the researcher being selective in the sampling, therefore by combining both questionnaires with interviews the study could have been strengthened. The tool of data collection used in the survey was post questionnaires to all maternity units in England and Wales. The questionnaires were addressed to the head of midwifery or senior midwife in each unit. The covering letter requested that the questionnaires were completed by the midwife with the most experience of postnatal depression (PND) services. This method possesses a number of problems within the information gathered for this article and for replicating the study at a later date. There is no guarantee that the most suitable person who completed the questionnaire did so in isolation or that it was given to the most appropriate person to complete. Some of the maternity units may have felt the need to promote their unit and to give the best impression, therefore there may have been a collaboration of heads that completed the questionnaires which could result in giving answers that do not demonstrate a completely true reflection. In other words this could be a biased response. The researchers sent out reminders after two weeks to units that had not responded to the questionnaires which in turn achieved an 86% response rate. There is no clear definition of what is an acceptable response rate but it is suggested that a response rate of 75%-85% from a postal survey is very good (Treece &Treece, 1986). However, there is no reference made to the non-respondents within the statistical information which suggests that a totally different result may have been given to the content of the article. It is proposed that it is important to know as much as possible about non-responders, as those who do not respond may differ from those who do and so introduce bias in the sampling (Moser & Kalton, 1971). Respondents could have been compared to non-respondents to establish similarity, so reducing the risk of bias. The final problem with the use of the questionnaires is the fact that the article pays no reference as to where to find the questions that were asked and therefore the quality of the questionnaire can not be analysed to see if it is written in a way to interest and motivate the respondents to reply positively. It is suggested that a questionnaire should not normally take longer than 20-25 minutes to complete (Treece & Treece, 1986). It is also suggested that questions should not be any longer than 20 words (Oppenheim, 1992). The use of quotes gives ‘life’ to the article and should provide support for the interpretation of the data (Cluett & Bluff, 2000). The quotes give the reader the impression that midwives are very much aware that there is a need for education and support systems within a multidisciplinary team with regards to referring cases of women with depression. It is perceived that this would enable midwives to carry out the role of screening for depression to a higher level. However, due to the anonymity of the source, the quotes may be questioned as there may be other reasons why the negative quotes are provided, for example there could be poor working relationship with the midwife and the units hierarchy. The sample size is composed of midwives employed in 211 maternity units in England and Wales. The size of the sample could be considered too small to be able to draw firm conclusions due to the fact that the sampling frame appears to be biased. The reason for this is the fact that the researchers selected only the midwives with the relevant knowledge of PND and guidelines for referral. This promotes some doubt as to whether the sample is appropriate which could adversely influence the validity and reliability of the study. It is necessary when planning the study that the researchers consider the characteristics that make individuals eligible for selection, and those that would make them unsuitable or even put them at risk or at a disadvantage (Rees, 2003). It is thought that the researchers would have an indication of the results therefore by excluding midwives lacking knowledge of the screening process and the inability to respond adequately this enhanced the results and provided justification for carrying out the study. There is no mention that a pilot study was carried out in order to check the consistency of the tool of data collection. This is probably mainly due to the fact that the qualitative researcher does not develop a standard tool of data collection derived from the literature review, or use a tool that has been validated in previous research. This enables the researcher to keep an open mind in important issues within the study, therefore it is not possible to carry out a pilot study as there is no structured research tool to test for reliability (Rees, 2003). Ethical considerations It is not clear whether ethical committee approval was gained. The ethical considerations apply to all research that involves the participant of clients and approval from the appropriate committee must be sought. Confidentiality and anonymity must be maintained and clear and concise information must be given to all participants to enable them the freedom from exploitation (Cluett & Bluff, 2000). As there appears to be no ethical approval for the study this can cause dilemmas, particularly as the sample was small. It would be unethical to undertake research on clients if the sample was too small to provide useful information (Cluett, 1996). Results The research results show that an 86% response rate was achieved from the questionnaires used to format the article and identifies the different response rates in regions from the lowest to the highest. No reason for further research has been suggested as to why there were such variations which ranged from 66% - 100%. The article provides information on 5 main domains being antenatal and postnatal depression screening, referral, information available to women, training and problems encountered by midwives. The article findings are shown in table form showing both percentages and the raw data so that the actual size of the denominator can be determined as this varies throughout the table due to respondents not completing all the questions asked of them. The results however do not appear to consider the amount of drop outs in the tests used and who undertakes the screening for antenatal depression as this falls from 182 respondents to 47 who answered questions. The results proceed to discuss what appears to show a distinct lack of specific training for midwives but as training is sometimes only felt as training when in a lecture type environment rather than on the job furthe...