- Email: [email protected]
Improving recordkeeping for maternal mortality programs, Kumasi, Ghana
International Journal of Gynecology & Obstetrics 59 Suppl. 2 Ž1997. S149]S155
Improving recordkeeping for maternal mortality programs, Kumasi, Ghana J.B. Danquaha,U , E.K. Appahb , J.O. Djanb , M. Ofori a , I.T. Essegbey a , S. Opokua , Ž The Kumasi PMM Team. a
b
Ministry of Health, Kumasi, Ghana Ministry of Health, School of Medical Sciences, Kumasi, Ghana
Abstract Preliminary studies: Data on obstetric complications are the basis of monitoring maternal mortality interventions in the PMM Network. A review of recordkeeping procedures at 10 facilities in the study area revealed that information on obstetric complications was often inconsistent or missing. Some hospital records were not designed to collect such information at all. Inter¨entions: In 1992, registers at facilities were revised to collect information on complications and time of treatment. Doctors, nurses, midwives and clerks were trained to record, compile and analyze data. Monitoring and supervisory mechanisms were also set up. Results: Recordkeeping has improved. Data collection and analysis have been regular and timely. Doctors have begun using the data for morning meetings. Nurses and midwives compile monthly summaries of data showing complications by type. Two other districts outside the research area have adopted the reporting system and it is possible that facilities in the whole region will follow suit. Costs: The cost of improving recordkeeping at the 10 healthcare facilities was approximately US $2543, with 85% coming from project funds. Conclusions: Existing recordkeeping systems can be modified to collect data necessary to monitor maternal mortality interventions. Staff training and monitoring visits are important to success. Q 1997 International Federation of Gynecology and Obstetrics Keywords: Africa; Ghana; Maternal mortality; Obstetric services; Records; Training; Cost
1. Introduction The launching of the Safe Motherhood Initiative in 1987 encouraged many international groups U
Corresponding author, Ministry of Health, Box 1908, Kumasi, Ghana.
to propose a variety of strategies to address the problem of maternal mortality. In the last five years, one of the big changes in the Initiative has been the increasing emphasis on obstetric complications as the key point of intervention. During that time, there has been a growing acceptance that, although most obstetric complications cannot be predicted or prevented, they can be treated.
0020-7292r97r$17.00 Q 1997 International Federation of Gynecology and Obstetrics PII S0020-7292Ž97. 00160-4
S150
J.B. Danquah et al. r International Journal of Gynecology & Obstetrics 59 Suppl. 2 (1997) S149]S155
Since 1988, 11 multidisciplinary teams from three West African countries ŽGhana, Nigeria and Sierra Leone. have focused on this aspect of the Initiative. This group, known as the Prevention of Maternal Mortality ŽPMM. Network has developed, implemented and evaluated strategies to prevent deaths from obstetric complications. They have received technical assistance from a team at Columbia University, New York. Their interventions have focused on improving the quality of, access to and utilization of emergency obstetric care. One important aspect of the Network’s program has been its emphasis on monitoring and improving data collection. Although obstetric complications are key events, they are not easy to identify in most existing record systems. Information about complications is often not recorded. When studies of particular complications have been done in hospitals, case notes have been used. The information they contain depends on the competency of the person who is writing them and the quality varies. Registers do exist in most health facilities, especially in the admission section of hospitals and health centers. However, in many cases, there is no column in the register which specifically asks whether a woman had a complication. Records provide visible evidence of what the hospital is accomplishing. When they are accurate and complete, they furnish a basis for evaluating hospital activities w1x. On the other hand, if they are inaccurate or incomplete, proper evaluation is difficult. For this reason, the PMM Network has greatly stressed recordkeeping, since achieving good data is necessary for properly evaluating research activities. Because the objective of the project is to improve the functioning and timely use of emergency obstetric care, even the effects of community mobilization can be partly monitored through facility records. One of the teams, based in Kumasi, Ghana, has concentrated particularly on improving the project area’s records and recordkeeping system. Many recording and data collection problems can be solved by having well-designed and accurately maintained registers. Because of this, the Kumasi
PMM team has channeled its efforts towards providing training and supervision in recordkeeping. 2. Preliminary studies Early in the project, the Kumasi PMM team embarked on a review of records and recordkeeping systems in 10 health facilities in their project area. The review was carried out over a fourmonth period from late 1991 to early 1992. Part of this review involved assessing whether the information required for the indicators used by PMM teams was being recorded w2x. The indicators used are the following: v v v
v
v v
number of obstetric admissions; number of deliveries; number of admissions for major obstetric complications Žoverall and by type.; number of maternal deaths Žoverall and by cause.; case fatality rate; and time from admission to treatment.
2.1. Methods Different research methods were used to identify problems areas. The Kumasi PMM team reviewed existing records and developed a checklist which it used to identify deficiencies and omissions in register and case notes. Existing procedures and information management systems were also observed. Finally, interviews were conducted with various levels of staff members to assess their knowledge of recordkeeping. 2.2. Findings The review of records and recordkeeping systems revealed deficiencies. Most of the information on obstetric complications in the registers was either lacking or inconsistent. Existing case notes were not designed to highlight information on complications. Information on the time of admission andror treatment was non-existent }
J.B. Danquah et al. r International Journal of Gynecology & Obstetrics 59 Suppl. 2 (1997) S149]S155
though most registers and case notes had information on dates. Records were poorly stored because of inadequate filing space. Problems in recordkeeping were also observed at the human resources and management level. A shortage of both trained records personnel and training programs was noted. Medical personnel lacked knowledge and skills in recording and managing information. Administrators’ underutilization of available information seemed to contribute to poor planning and management of services. 3. Interventions A number of interventions were instituted by the Kumasi PMM team in late 1992 and 1993, to correct the deficiencies identified. These included training, improving registers and monitoring and supervising recordkeeping. The activities related to each of these areas are described more fully below.
S151
team would have baseline data available when they wanted to evaluate their project. Records department clerks, who were previously untrained, were instructed on how to identify, compile and analyze data necessary for maternal mortality indicators. Nurses and midwives were trained to record data for these indicators and to keep proper records in the wards. Doctors and medical assistants were trained in recording information on patient diagnoses and treatment. Doctors and medical assistants were also taught how to review and use collected data to improve patient management and service delivery. For example, a time]motion study done at the teaching hospital revealed avoidable delays in the admission-to-treatment interval w3x. One such delay was caused by a lack of ready access to emergency drugs. To address this, a 24-h pharmacy system was instituted. The various categories of personnel trained are summarized in Table 1. 3.2. Impro¨ing registers
3.1. Training A series of training programs was carried out to update the knowledge and skills of personnel handling records in the project area. Ministry of Health ŽMOH. staff working in 10 district facilities, as well as those from Komfo Anokye Teaching Hospital in Kumasi, participated in the training. These training programs were intentionally held before the PMM team’s other service-related interventions were implemented, so that the
In the Kumasi project area, as in many maternity facilities in West Africa, the basic registers that can be found are the following: v v
v
the theater register in the operating theater; the general admissions and discharge register Žknown as the nominal roll in Ghana. for the entire facility, which is kept in the records department; the delivery register in the labor ward; and
Table 1 Personnel trained in recordkeeping ŽKumasi, Ghana, 1992. Staff level
Health facility
Number trained
Duration of training
Record clerks
KATH Project districts KATH Project districts KATH Project districts Project districts
30 25 40 30 16 4 15
3]4 days 3]4 days 2]7 days 2]7 days 1]5 days 1]5 days 1]5 days
Nurses and midwives Doctors Medical assistants
KATH, Komfo Anokye Teaching Hospital.
S152 v
J.B. Danquah et al. r International Journal of Gynecology & Obstetrics 59 Suppl. 2 (1997) S149]S155
the admission and discharge ŽA and D. register on every ward.
Procedures for recording information and in some cases the data collection forms themselves, were modified in most facilities. The team observed that the theater register was the only register that contained accurate information on complications, so no changes were made there. The general admission and discharge register in the records department had a column for recording complications, but this was not consistently filled out. In the delivery register, it was noted that the staff did not complete some of the columns, especially the column for pregnancy complications. This register was sometimes used for recording normal deliveries only. In the nominal roll register provided by the MOH, additional columns for describing the condition on discharge and type of treatment were added. Until the team’s intervention, the nominal roll had been used only in hospitals. The PMM team found it to be useful to the staff of health centers as well. Upon the team’s recommendation, it was introduced at that level in the project district. Fig. 1 illustrates the column headings of the revised nominal roll. A lack of information on complications was also observed in the maternity wards: the A and D and delivery registers on the maternity wards did not have columns to record information on complications. In addition, most of these registers were in a dilapidated condition. This prompted
the team to design and provide the maternity wards at each institution with new sets of A and D registers. We redesigned the delivery and records department registers to reflect information on complications and time of treatment. The A and D registers were modified to include information on the following: initial diagnosis; final diagnosis; date and time of admission; date and time of treatment; and place from which the patient was referred Žsee Fig. 2.. The initial and final diagnosis columns of the maternity A and D register were included in case a patient admitted with one diagnosis ended up developing a major obstetric complication. In the event of a maternal death, the word ‘died’ was entered under the column titled ‘condition on discharge’. Columns for date and time of admission and date and time of treatment were also added. From these data, the promptness with which a patient was treated after being admitted to the hospital could be measured. In addition to completeness, consistency was also noted as being important. For example, with the old system, information in the columns of a page in the register would be lost if the person ruling the succeeding page were not conscientious in transferring the same headings to the next page. To improve the consistency with which information was recorded, one inch was cut off the tops of the pages of these registers and the column headings were written on slips of paper and fixed to the top of the back cover of the register. This ensured that the column headings were consis-
Fig. 1. Nominal roll for maternity patients Žkept in records department for recording daily admissions..
Fig. 2. Admission and discharge register Žfor maternity wards, daily recording..
J.B. Danquah et al. r International Journal of Gynecology & Obstetrics 59 Suppl. 2 (1997) S149]S155
tent. It also helped save time and avoided the mixing up of information. 3.3. Monitoring and super¨ision To promote the implementation and maintenance of the improved recordkeeping system,
S153
monitoring and supervisory mechanisms were set up. A checklist, as shown in Fig. 3, was designed for use by the PMM team, the District Health Management Team ŽDHMT. and individual officers who make supervisory visits to the health institutions. This checklist is used both for monthly and quarterly monitoring rounds. Indi-
Fig. 3. Sample monitoring and supervisory checklist, Kumasi, Ghana.
S154
J.B. Danquah et al. r International Journal of Gynecology & Obstetrics 59 Suppl. 2 (1997) S149]S155
vidual district officers make monthly visits, while the PMM team and the DHMT make quarterly supervisory visits. District medical officers were brought into the team from the onset and they joined in the planning of activities. This helped the staff at the health institutions to not regard the PMM team as intruders. Instead, medical officers, senior nurses and midwives participated in the monitoring of activities at the institution and district levels. Feedback about research activities was periodically sent in report form to policymakers at the regional and national levels. 4. Results The Kumasi PMM Research Project has brought considerable improvement to recordkeeping at the two district and the tertiary hospitals. Available training reports, as well as observations and data from supervisory visits, indicate that the staff have acquired relevant skills and are more confident in their recordkeeping. Record clerks submit midwives’ monthly reports to the regional health administration regularly and on time. Nurses and midwives make monthly summaries of data in the ward register showing complications by type. The nominal roll is filled in duplicate and the original is sent to the regional health administration, as required by the MOH, on a monthly basis. Prior to the project, reports generally were not submitted on time. Record clerks also provide analyses of quarterly and annual data showing trends. They then disseminate this information to medical personnel and administrators for the preparation of reports. Doctors at the teaching hospital now have morning meetings daily to review the previous day’s cases using the data they have recorded. Improvements brought about by this project have also had spillover effects. The record clerks in the project area have become a source of inspiration and encouragement for record clerks in two other districts. Clerks in the other districts have requested that their district medical officers arrange for their training. The regional biostatistics division plans to incorporate the PMM training into their regional training program, which
will be organized by the MOH. There has also been a great improvement in relations among different units of the health institutions. The records departments have been receiving needed cooperation because the flow of information and data is more consistent and its content more accurate. In two other districts outside the research area, it has now become routine to report the places from which women were referred. These districts report not only the maternity cases, but general emergencies as well. From the interest being expressed, this system might soon cover the whole region. Finally, the Kumasi PMM team has also forged links with other governmental and non-governmental entities. One important lesson the research team learned is that the success of the intervention was greatly due to the involvement, from the onset, of MOH personnel at the district, regional and even national level. The team also had the pleasure and the privilege to receive visits from others wishing to learn from our project. The Safe Motherhood Program Officer from UNICEF Ghana visited the team in Kumasi to inspect and assess the recordkeeping system that the team had put in place. As a result of her visit, UNICEF sponsored a regional workshop by UNICEF for health workers to spread word of the Kumasi PMM team’s recordkeeping system. 5. Cost data The cost of this intervention was mostly borne by the Kumasi PMM team’s project grant from the Carnegie Corporation, with assistance from the MOH. As Table 2 shows, the total cost of the intervention Ž$2543., when divided among the 10 upgraded institutions in the project district, amounts to $254 per institution. The MOH contribution Žconsisting of registers and stationery. was 15% of the total cost, which complemented the 85% donated by the project. It should be noted that the 15% contribution made by the MOH only represents the additional costs of these activities and not ongoing expenses such as salaries of existing staff.
J.B. Danquah et al. r International Journal of Gynecology & Obstetrics 59 Suppl. 2 (1997) S149]S155
S155
Table 2 Cost of upgrading recordkeeping systems in project districts and KATH, Kumasi, Ghana, 1992]1993 Item
Cost ŽUS $.
Percent of of total Ž%.
Cost to whom Žpercent of component. Project
MOH
Training and supervision Registers Stationery Printing of referral forms
1667 639 161 76
66 25 6 3
100 52 69 74
0 48 31 26
Total
2543
100
85
15
6. Discussion Recordkeeping systems for monitoring maternal mortality programs can improve considerably if the necessary mechanisms are put in place and supervised. These improvements can be accomplished using Žor slightly modifying. existing recordkeeping systems and without great expense or technological requirements. Staff training is an invaluable component of this process. The monthly and quarterly supervising and monitoring activities of district officers, DHMT and the PMM team have put the staff at the health facilities on the alert and has increased their desire to improve their performance. Cooperation across staff levels and management’s support for the records unit are also important. Data generated on obstetric complications are crucial for managing and evaluating the prevention of maternal mortality services. However, for statistical data to be of real value, they must be accurate. They can only be accurate if the records from which the figures are gathered are themselves accurate w4x. The work done by the Kumasi team demonstrates that efforts to improve recordkeeping may provide a low cost, low technology way to document progress toward safe motherhood.
Acknowledgements The Kumasi PMM Team is extremely grateful to the Carnegie Corporation of New York for funding this project. We would also like to thank Columbia University and the Columbia PMM team, especially Deborah Maine, Angela Kamara, Jennifer Brown and Therese McGinn, for their technical support and encouragement. We wish to express our sincere gratitude to the following people for their contributions: J.O. Martey, S.W.K. Adadevoh, Charles Sereboo, Abel Botchway and Cecilia Manu. References w1x Huffman E. Manual for medical record librarians. 4th ed. Berwyn, Illinois: Physician’s Company, 1960:31]107. w2x Prevention of Maternal Mortality Network. Second Operations Research Methods Workshop, Including Focus Group Discussions. Developed during Research Methods Workshop held in Enugu, Nigeria, December 12]16, 1988. w3x Kumasi PMM Team. Final report, 1991:39]44 Žunpublished.. w4x Huffman E. Manual for medical record librarians. 4th ed. Berwyn, Illinois: Physician’s Company, 1960:310.
Improving recordkeeping for maternal mortality programs, Kumasi, Ghana J.B. Danquaha,U , E.K. Appahb , J.O. Djanb , M. Ofori a , I.T. Essegbey a , S. Opokua , Ž The Kumasi PMM Team. a
b
Ministry of Health, Kumasi, Ghana Ministry of Health, School of Medical Sciences, Kumasi, Ghana
Abstract Preliminary studies: Data on obstetric complications are the basis of monitoring maternal mortality interventions in the PMM Network. A review of recordkeeping procedures at 10 facilities in the study area revealed that information on obstetric complications was often inconsistent or missing. Some hospital records were not designed to collect such information at all. Inter¨entions: In 1992, registers at facilities were revised to collect information on complications and time of treatment. Doctors, nurses, midwives and clerks were trained to record, compile and analyze data. Monitoring and supervisory mechanisms were also set up. Results: Recordkeeping has improved. Data collection and analysis have been regular and timely. Doctors have begun using the data for morning meetings. Nurses and midwives compile monthly summaries of data showing complications by type. Two other districts outside the research area have adopted the reporting system and it is possible that facilities in the whole region will follow suit. Costs: The cost of improving recordkeeping at the 10 healthcare facilities was approximately US $2543, with 85% coming from project funds. Conclusions: Existing recordkeeping systems can be modified to collect data necessary to monitor maternal mortality interventions. Staff training and monitoring visits are important to success. Q 1997 International Federation of Gynecology and Obstetrics Keywords: Africa; Ghana; Maternal mortality; Obstetric services; Records; Training; Cost
1. Introduction The launching of the Safe Motherhood Initiative in 1987 encouraged many international groups U
Corresponding author, Ministry of Health, Box 1908, Kumasi, Ghana.
to propose a variety of strategies to address the problem of maternal mortality. In the last five years, one of the big changes in the Initiative has been the increasing emphasis on obstetric complications as the key point of intervention. During that time, there has been a growing acceptance that, although most obstetric complications cannot be predicted or prevented, they can be treated.
0020-7292r97r$17.00 Q 1997 International Federation of Gynecology and Obstetrics PII S0020-7292Ž97. 00160-4
S150
J.B. Danquah et al. r International Journal of Gynecology & Obstetrics 59 Suppl. 2 (1997) S149]S155
Since 1988, 11 multidisciplinary teams from three West African countries ŽGhana, Nigeria and Sierra Leone. have focused on this aspect of the Initiative. This group, known as the Prevention of Maternal Mortality ŽPMM. Network has developed, implemented and evaluated strategies to prevent deaths from obstetric complications. They have received technical assistance from a team at Columbia University, New York. Their interventions have focused on improving the quality of, access to and utilization of emergency obstetric care. One important aspect of the Network’s program has been its emphasis on monitoring and improving data collection. Although obstetric complications are key events, they are not easy to identify in most existing record systems. Information about complications is often not recorded. When studies of particular complications have been done in hospitals, case notes have been used. The information they contain depends on the competency of the person who is writing them and the quality varies. Registers do exist in most health facilities, especially in the admission section of hospitals and health centers. However, in many cases, there is no column in the register which specifically asks whether a woman had a complication. Records provide visible evidence of what the hospital is accomplishing. When they are accurate and complete, they furnish a basis for evaluating hospital activities w1x. On the other hand, if they are inaccurate or incomplete, proper evaluation is difficult. For this reason, the PMM Network has greatly stressed recordkeeping, since achieving good data is necessary for properly evaluating research activities. Because the objective of the project is to improve the functioning and timely use of emergency obstetric care, even the effects of community mobilization can be partly monitored through facility records. One of the teams, based in Kumasi, Ghana, has concentrated particularly on improving the project area’s records and recordkeeping system. Many recording and data collection problems can be solved by having well-designed and accurately maintained registers. Because of this, the Kumasi
PMM team has channeled its efforts towards providing training and supervision in recordkeeping. 2. Preliminary studies Early in the project, the Kumasi PMM team embarked on a review of records and recordkeeping systems in 10 health facilities in their project area. The review was carried out over a fourmonth period from late 1991 to early 1992. Part of this review involved assessing whether the information required for the indicators used by PMM teams was being recorded w2x. The indicators used are the following: v v v
v
v v
number of obstetric admissions; number of deliveries; number of admissions for major obstetric complications Žoverall and by type.; number of maternal deaths Žoverall and by cause.; case fatality rate; and time from admission to treatment.
2.1. Methods Different research methods were used to identify problems areas. The Kumasi PMM team reviewed existing records and developed a checklist which it used to identify deficiencies and omissions in register and case notes. Existing procedures and information management systems were also observed. Finally, interviews were conducted with various levels of staff members to assess their knowledge of recordkeeping. 2.2. Findings The review of records and recordkeeping systems revealed deficiencies. Most of the information on obstetric complications in the registers was either lacking or inconsistent. Existing case notes were not designed to highlight information on complications. Information on the time of admission andror treatment was non-existent }
J.B. Danquah et al. r International Journal of Gynecology & Obstetrics 59 Suppl. 2 (1997) S149]S155
though most registers and case notes had information on dates. Records were poorly stored because of inadequate filing space. Problems in recordkeeping were also observed at the human resources and management level. A shortage of both trained records personnel and training programs was noted. Medical personnel lacked knowledge and skills in recording and managing information. Administrators’ underutilization of available information seemed to contribute to poor planning and management of services. 3. Interventions A number of interventions were instituted by the Kumasi PMM team in late 1992 and 1993, to correct the deficiencies identified. These included training, improving registers and monitoring and supervising recordkeeping. The activities related to each of these areas are described more fully below.
S151
team would have baseline data available when they wanted to evaluate their project. Records department clerks, who were previously untrained, were instructed on how to identify, compile and analyze data necessary for maternal mortality indicators. Nurses and midwives were trained to record data for these indicators and to keep proper records in the wards. Doctors and medical assistants were trained in recording information on patient diagnoses and treatment. Doctors and medical assistants were also taught how to review and use collected data to improve patient management and service delivery. For example, a time]motion study done at the teaching hospital revealed avoidable delays in the admission-to-treatment interval w3x. One such delay was caused by a lack of ready access to emergency drugs. To address this, a 24-h pharmacy system was instituted. The various categories of personnel trained are summarized in Table 1. 3.2. Impro¨ing registers
3.1. Training A series of training programs was carried out to update the knowledge and skills of personnel handling records in the project area. Ministry of Health ŽMOH. staff working in 10 district facilities, as well as those from Komfo Anokye Teaching Hospital in Kumasi, participated in the training. These training programs were intentionally held before the PMM team’s other service-related interventions were implemented, so that the
In the Kumasi project area, as in many maternity facilities in West Africa, the basic registers that can be found are the following: v v
v
the theater register in the operating theater; the general admissions and discharge register Žknown as the nominal roll in Ghana. for the entire facility, which is kept in the records department; the delivery register in the labor ward; and
Table 1 Personnel trained in recordkeeping ŽKumasi, Ghana, 1992. Staff level
Health facility
Number trained
Duration of training
Record clerks
KATH Project districts KATH Project districts KATH Project districts Project districts
30 25 40 30 16 4 15
3]4 days 3]4 days 2]7 days 2]7 days 1]5 days 1]5 days 1]5 days
Nurses and midwives Doctors Medical assistants
KATH, Komfo Anokye Teaching Hospital.
S152 v
J.B. Danquah et al. r International Journal of Gynecology & Obstetrics 59 Suppl. 2 (1997) S149]S155
the admission and discharge ŽA and D. register on every ward.
Procedures for recording information and in some cases the data collection forms themselves, were modified in most facilities. The team observed that the theater register was the only register that contained accurate information on complications, so no changes were made there. The general admission and discharge register in the records department had a column for recording complications, but this was not consistently filled out. In the delivery register, it was noted that the staff did not complete some of the columns, especially the column for pregnancy complications. This register was sometimes used for recording normal deliveries only. In the nominal roll register provided by the MOH, additional columns for describing the condition on discharge and type of treatment were added. Until the team’s intervention, the nominal roll had been used only in hospitals. The PMM team found it to be useful to the staff of health centers as well. Upon the team’s recommendation, it was introduced at that level in the project district. Fig. 1 illustrates the column headings of the revised nominal roll. A lack of information on complications was also observed in the maternity wards: the A and D and delivery registers on the maternity wards did not have columns to record information on complications. In addition, most of these registers were in a dilapidated condition. This prompted
the team to design and provide the maternity wards at each institution with new sets of A and D registers. We redesigned the delivery and records department registers to reflect information on complications and time of treatment. The A and D registers were modified to include information on the following: initial diagnosis; final diagnosis; date and time of admission; date and time of treatment; and place from which the patient was referred Žsee Fig. 2.. The initial and final diagnosis columns of the maternity A and D register were included in case a patient admitted with one diagnosis ended up developing a major obstetric complication. In the event of a maternal death, the word ‘died’ was entered under the column titled ‘condition on discharge’. Columns for date and time of admission and date and time of treatment were also added. From these data, the promptness with which a patient was treated after being admitted to the hospital could be measured. In addition to completeness, consistency was also noted as being important. For example, with the old system, information in the columns of a page in the register would be lost if the person ruling the succeeding page were not conscientious in transferring the same headings to the next page. To improve the consistency with which information was recorded, one inch was cut off the tops of the pages of these registers and the column headings were written on slips of paper and fixed to the top of the back cover of the register. This ensured that the column headings were consis-
Fig. 1. Nominal roll for maternity patients Žkept in records department for recording daily admissions..
Fig. 2. Admission and discharge register Žfor maternity wards, daily recording..
J.B. Danquah et al. r International Journal of Gynecology & Obstetrics 59 Suppl. 2 (1997) S149]S155
tent. It also helped save time and avoided the mixing up of information. 3.3. Monitoring and super¨ision To promote the implementation and maintenance of the improved recordkeeping system,
S153
monitoring and supervisory mechanisms were set up. A checklist, as shown in Fig. 3, was designed for use by the PMM team, the District Health Management Team ŽDHMT. and individual officers who make supervisory visits to the health institutions. This checklist is used both for monthly and quarterly monitoring rounds. Indi-
Fig. 3. Sample monitoring and supervisory checklist, Kumasi, Ghana.
S154
J.B. Danquah et al. r International Journal of Gynecology & Obstetrics 59 Suppl. 2 (1997) S149]S155
vidual district officers make monthly visits, while the PMM team and the DHMT make quarterly supervisory visits. District medical officers were brought into the team from the onset and they joined in the planning of activities. This helped the staff at the health institutions to not regard the PMM team as intruders. Instead, medical officers, senior nurses and midwives participated in the monitoring of activities at the institution and district levels. Feedback about research activities was periodically sent in report form to policymakers at the regional and national levels. 4. Results The Kumasi PMM Research Project has brought considerable improvement to recordkeeping at the two district and the tertiary hospitals. Available training reports, as well as observations and data from supervisory visits, indicate that the staff have acquired relevant skills and are more confident in their recordkeeping. Record clerks submit midwives’ monthly reports to the regional health administration regularly and on time. Nurses and midwives make monthly summaries of data in the ward register showing complications by type. The nominal roll is filled in duplicate and the original is sent to the regional health administration, as required by the MOH, on a monthly basis. Prior to the project, reports generally were not submitted on time. Record clerks also provide analyses of quarterly and annual data showing trends. They then disseminate this information to medical personnel and administrators for the preparation of reports. Doctors at the teaching hospital now have morning meetings daily to review the previous day’s cases using the data they have recorded. Improvements brought about by this project have also had spillover effects. The record clerks in the project area have become a source of inspiration and encouragement for record clerks in two other districts. Clerks in the other districts have requested that their district medical officers arrange for their training. The regional biostatistics division plans to incorporate the PMM training into their regional training program, which
will be organized by the MOH. There has also been a great improvement in relations among different units of the health institutions. The records departments have been receiving needed cooperation because the flow of information and data is more consistent and its content more accurate. In two other districts outside the research area, it has now become routine to report the places from which women were referred. These districts report not only the maternity cases, but general emergencies as well. From the interest being expressed, this system might soon cover the whole region. Finally, the Kumasi PMM team has also forged links with other governmental and non-governmental entities. One important lesson the research team learned is that the success of the intervention was greatly due to the involvement, from the onset, of MOH personnel at the district, regional and even national level. The team also had the pleasure and the privilege to receive visits from others wishing to learn from our project. The Safe Motherhood Program Officer from UNICEF Ghana visited the team in Kumasi to inspect and assess the recordkeeping system that the team had put in place. As a result of her visit, UNICEF sponsored a regional workshop by UNICEF for health workers to spread word of the Kumasi PMM team’s recordkeeping system. 5. Cost data The cost of this intervention was mostly borne by the Kumasi PMM team’s project grant from the Carnegie Corporation, with assistance from the MOH. As Table 2 shows, the total cost of the intervention Ž$2543., when divided among the 10 upgraded institutions in the project district, amounts to $254 per institution. The MOH contribution Žconsisting of registers and stationery. was 15% of the total cost, which complemented the 85% donated by the project. It should be noted that the 15% contribution made by the MOH only represents the additional costs of these activities and not ongoing expenses such as salaries of existing staff.
J.B. Danquah et al. r International Journal of Gynecology & Obstetrics 59 Suppl. 2 (1997) S149]S155
S155
Table 2 Cost of upgrading recordkeeping systems in project districts and KATH, Kumasi, Ghana, 1992]1993 Item
Cost ŽUS $.
Percent of of total Ž%.
Cost to whom Žpercent of component. Project
MOH
Training and supervision Registers Stationery Printing of referral forms
1667 639 161 76
66 25 6 3
100 52 69 74
0 48 31 26
Total
2543
100
85
15
6. Discussion Recordkeeping systems for monitoring maternal mortality programs can improve considerably if the necessary mechanisms are put in place and supervised. These improvements can be accomplished using Žor slightly modifying. existing recordkeeping systems and without great expense or technological requirements. Staff training is an invaluable component of this process. The monthly and quarterly supervising and monitoring activities of district officers, DHMT and the PMM team have put the staff at the health facilities on the alert and has increased their desire to improve their performance. Cooperation across staff levels and management’s support for the records unit are also important. Data generated on obstetric complications are crucial for managing and evaluating the prevention of maternal mortality services. However, for statistical data to be of real value, they must be accurate. They can only be accurate if the records from which the figures are gathered are themselves accurate w4x. The work done by the Kumasi team demonstrates that efforts to improve recordkeeping may provide a low cost, low technology way to document progress toward safe motherhood.
Acknowledgements The Kumasi PMM Team is extremely grateful to the Carnegie Corporation of New York for funding this project. We would also like to thank Columbia University and the Columbia PMM team, especially Deborah Maine, Angela Kamara, Jennifer Brown and Therese McGinn, for their technical support and encouragement. We wish to express our sincere gratitude to the following people for their contributions: J.O. Martey, S.W.K. Adadevoh, Charles Sereboo, Abel Botchway and Cecilia Manu. References w1x Huffman E. Manual for medical record librarians. 4th ed. Berwyn, Illinois: Physician’s Company, 1960:31]107. w2x Prevention of Maternal Mortality Network. Second Operations Research Methods Workshop, Including Focus Group Discussions. Developed during Research Methods Workshop held in Enugu, Nigeria, December 12]16, 1988. w3x Kumasi PMM Team. Final report, 1991:39]44 Žunpublished.. w4x Huffman E. Manual for medical record librarians. 4th ed. Berwyn, Illinois: Physician’s Company, 1960:310.