International Journal for Quality in Health Care Advance Access originally published online on April 14, 2005
International Journal for Quality in Health Care 2005 17(4):293-300; doi:10.1093/intqhc/mzi042
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Quality improvement programme on the frontline: An International Committee of the Red Cross experience in the Democratic Republic of Congo
ICRC, Medical Unit, Genève, Switzerland
Background. Majority of research in Quality Improvement, focuses on developed countries or development programs. Humanitarian organisations frequently work in developing countries, often in emergency situations with rapid staff turnover. Objectives of this study are twofold: first to develop a methodology of motivation and restoration of collapsed health structures through the creation of community based QI indicators; second, to implement these indicators to improve quality of care.
Methodology. Using a community-based approach, the International Committee of the Red Cross (ICRC) team together with local health committee and nurses developed quality indicators divided into six different categories. Of 16 community primary health centers and four hospitals supported by ICRC, six health centers and one hospital were chosen to follow quality indicators for three of six indicator categories. Initial data were collected in January 2003 and compared with data serially gathered throughout the year.
Results. In the category rational prescription, all health facilities except for one showed improvement in every category. In the hygiene category, four of seven health structures showed 100% improvement in their score. Three of seven facilities showed impressive improvement in the category pharmacy management.
Conclusion. Involving the community to design population based indicators helped communities take ownership of the indicators. Our findings that poor performance on indicators prompted communities to seek training and assistance to improve quality of care emphasized this. Continued adherence and improvement in each category confirmed the long term effects of teaching sessions in the areas of rational prescription, hygiene and pharmacy maintenance.
Keywords: quality indicators, Red Cross, supervision
Address reprint requests to Stéphane du Mortier, ICRC, Medical Unit, 19 Avenue de la Paix, 1202 Genève, Switzerland. E-mail: stephdumortier{at}hotmail.com
Accepted for publication March 6, 2005.
In the health care sector, the necessity of following quality of care through use of quality improvement tools is well known and documented [1–4]. The majority of research, however, has focused on developed countries or in development programmes. Humanitarian organizations frequently work in developing countries, their work often in emergency situations with rapid staff turnover, in settings which are frequently volatile [5]. If internal quality improvement reports are written, few are published [6,7].
The International Committee of the Red Cross (ICRC) has been present in the eastern provinces of the Democratic Republic of Congo (DRC) since 1993. As part of the overall ICRC team in the DRC, the medical staff work in conjunction with local and expatriate colleagues to fulfil the ICRC mandate of providing protection and assistance along current or former front lines to victims of armed conflict. The objective of ICRC medical assistance in DRC is to provide quality health care through uninterrupted support to local health care facilities [8]. Methods of achieving this goal include maintenance of existing health structures, proper management of human resources and materials, and introduction, training, and monitoring of quality of care. Implementation is realized through organization of regular training sessions and monthly supervision in conjunction with deliveries of medication and medical supplies.
In 2003, the ICRC medical team assisted 16 rural community primary health care centres and four referral hospitals in five provinces in eastern DRC. As the ICRC-supported health structures are community based, participation of the community in the decision-making process was a priority. Although several articles indicate the importance of community involvement in the appraisal of health care priorities [9,10], no additional literature was found which included the effect of community involvement in the creation of quality improvement indicators.
As a member of the Active Learning Network of Accountability and Performance in Humanitarian Action (ALNAP) [11], the ICRC looks for methods to quantify, monitor, and document quality. While it has been documented that humanitarian organizations have begun to analyse the quality of their work [11–14], information regarding quality improvement programmes in humanitarian emergency or front-line settings is not published or otherwise available for review. We performed this study, therefore, with two goals: first, to develop a methodology of motivation and restoration of collapsed health structures through the creation of community-based quality improvement indicators, and secondly, to implement these indicators to improve quality of care.
 |
Setting and participants |
|---|
 Top Setting and participants Development of methodologyAnalysisResultsDiscussionLimitationsImplicationsReferences |
|---|
Of 16 rural health centres and four referral hospitals ICRC supported in 2003, six health centres in the Equator province (eastern Ikela) and one referral hospital in the Katanga province (Kalemie) were chosen for this study. Selection was based on early introduction of indicators, availability of data, and early acceptance of the ICRC programme by staff. Although supported facilities are situated in five different provinces, the most common patient pathologies within each health structure remain the same: malaria, respiratory tract infections, diarrhoeal pathologies, sexually transmitted diseases, and measles.
While all ICRC supported health centres in the DRC are in remote village settings, the clinics chosen for this study are the most remote, accessible only by a 3-day motorcycle ride on dirt jungle paths from Kisangani. The health centres are located 15–50 kilometers from former front lines closed to population traffic. As with other ICRC-supported centres, a local ICRC physician is placed in the area to provide monthly supervision and training.
In the DRC, each health centre is comprised of two components: the medical team and the local health committee (named COSA). The medical team consists of a nurse and a general assistant. The health committee comprises 18 elected representatives of the villages served by the health centre. While the primary task of the medical team is to provide health care to the served population, the health committee acts as the liaison between health centre and population. This committee also monitors administration of the health centre.
The private hospital chosen for this study is a 120-bed reference hospital in the northern part of the Katanga province. All administrative decisions are made in conjunction with the railroad company (SNCC) that owns the hospital; medical and nursing decisions are made by the head nurse of each respective department. The outpatient department (OPD) encounters the same pathologies seen at all community health centres.
|
Development of methodology |
|---|
|
TopSetting and participantsDevelopment of methodologyAnalysisResultsDiscussionLimitationsImplicationsReferences |
|---|
After review of available literature concerning methodology of quality improvement [1,15] we developed the following step-by-step process. All steps were a collaborative effort developed by health committee representatives, health centre staff, and the local ICRC doctor.
Step one: identify community priorities
After two to three community meetings, stakeholders agreed upon community priorities for the health centre. These ranged from structural priorities (e.g. building or toilet facility renovation) to qualitative ones (e.g. cleanliness of facility, rational use of medication, organization of pharmacy).
Step two: analyse and translate priorities to quarterly objectives
Using regional and national governmental norms as a reference, two representatives of the health committee, with the local nurse and the assistance of the ICRC doctor, transformed the priorities into objectives. Agreed objectives were then shared with the community.
Step three: develop and test at least one indicator for every objective
This and the previous step were reached in the same day. Through questions asked by the ICRC doctor, such as ‘How will we agree we achieved the objectives?’ and ‘What indicator would allow us to check whether we reached our objectives?’, the need arose for agreed-upon indicators to quantify and verify that the objective was achieved.
For structural objectives, indicators were relatively easy to agree upon. For example, the general objective ‘to provide proper hygiene in the centre’ was translated to the specific objective ‘a toilet needs to be built’ and the indicator became ‘a clean latrine should be available at least 20 meters from the health centre building’. However, with more qualitative objectives such as ‘having an adequate supply of medicine to treat all patients’, indicators such as ‘no interruption of medication during the quarter’ were more difficult to agree upon. For rational prescriptions, ICRC proposed the key indicators developed by the World Health Organization [16].
While the first three steps were planned, the process of translating objectives to specific indicators proved frustrating for the health committee and health centre members, necessitating the development of the following two strategies by the ICRC doctor. Two to three days per health centre were dedicated each month to assist in this task. Quarterly meetings were held with all the health committee members and health centre staff from the same region for common training and collaboration. Unfortunately, because of security problems these appointments could not always take place as planned.
Step four: analysing general objectives to develop specific indicators
Five months after initiating the process of developing the above steps, the health committee, during one of the regional meetings, asked for assistance in translating general objectives into specific objectives and linking indicators to these. For example, ‘performing a better consultation’ was acknowledged to be a succession of steps ranging from reception of the patient to delivery of medication. Thus, every step (e.g. taking patient history, physical examination, prescribing laboratory tests) was considered a source of specific indicators.
Step five: development of indicator categories
With time, all collaborative partners realized that several identified specific indicators could be placed into general categories (Table 1). The six indicator categories created in 2002 and 2003 by the health committee, health committee and ICRC teams included rational prescriptions, pharmacy management, hygiene, quality of the committee itself, water source management and quality of the immunization programme.
|
The following steps were subsequently created in order to monitor and document progress.
Step six: data collection for monitoring and creation of the point system
Of the six identified indicator categories, three were selected for close monitoring in 2003: rational prescriptions, pharmacy management, and hygiene. During each unannounced ICRC quarterly visit, health centre staff joined the ICRC doctor to analyse each indicator. Points were awarded accordingly. The score became a consensus between three or four people. The pharmacy management indicator allowed us to gather data to assess the management of the pharmacy stock and follow consumption of each medication in order to avoid misuse or theft. This category was important for the accountability of health centre staff in front of the health committee.
These categories were chosen as they were the first to be introduced and implemented at the end of 2002. Points were attributed to each indicator in order to quantify results (Table 2).
|
Step seven: presentation of data to different partners
In 2003, data were collected and quarterly feedback presented to the community by the health committee, with health centre staff and the ICRC doctor present. During these meetings, we discovered the fascination of the local communities for outside evaluation of their performance and the subsequent results. It became a game for them to see how many points they could attain, to the point that they would dispute each point taken away.
|
Analysis |
|---|
|
TopSetting and participantsDevelopment of methodologyAnalysisResultsDiscussionLimitationsImplicationsReferences |
|---|
We chose a prospective, descriptive study design where all selected indicators were checked quarterly. ICRC doctors, collecting data during unannounced quarterly visits, provided follow-up comparison for each health structure. In the rational prescription category, findings were compared against national norms. Chi-squared testing was performed for the average number of drugs to look for significance in improvement. In all categories, we considered any improvement in score good. Attainment and maintenance of scores better than the norm was considered excellent.
|
Results |
|---|
|
TopSetting and participantsDevelopment of methodologyAnalysisResultsDiscussionLimitationsImplicationsReferences |
|---|
Rational prescription category
Comparison of initial and final data (January to December) (Table 3) indicates that all health facilities with the exception of Anzi (percentage of encounters with an injection) showed improvement in every category. With the indicator ‘average number of drugs prescribed per encounter’, everyone had either reached or outperformed the norm except Mpona (high prevalence of STDs with all prescriptions respecting national protocols) and showed statistically significant improvement. By the end of 2003, all health structures were using 100% generic drugs except for Bosango which was still within the norm. All showed marked improvement with antibiotic prescriptions and were within the norm, except Monkoso (high prevalence of sexually transmitted diseases—STDs with all prescriptions respecting national protocols). Results with the indicator ‘percentage of encounters with an injection’ again showed overall improvement, with only Anzi remaining above the norm (no epidemiological justification). The indicator used to detect prescription contraindications showed improvement with all health structures prescribing techniques, although three health structures did not meet the excellence norm (zero).
|
Hygiene indicator category
Four of the seven health structures showed 100% improvement in their score. While Lofusoloko started out with 7 of 10 points, their improvement to 9 points, while not remarkable, was considered an improvement. Only Bosango and Kalemie indicated a static state from the beginning to the year’s end.
Pharmacy management category
Unfortunately, initial data were not available for two of seven facilities (Ikamoloki and Mpona) as they were forced to hide their stock due to regional insecurity posing a threat of theft. Three of seven facilities showed impressive improvement, notably Anzi and Lofusoloko, who initially received no points, to Lofusoloko scoring the maximum number of points by the end of 2003. Again, Bosango and Kalemie showed a steady state from the beginning to the year’s end.
|
Discussion |
|---|
|
TopSetting and participantsDevelopment of methodologyAnalysisResultsDiscussionLimitationsImplicationsReferences |
|---|
Steady improvement, as shown by the data, indicated the motivation of the supported health facilities to improve. Additionally, their continued requests for supervision and training emphasized this point. We attributed the success of this programme to the community involvement we sought from the beginning which aided in their willingness to claim ownership of the indicators.
Mpona and Monkoso, while respecting national protocols, continued to be outside the norm for the indicators ‘average number of drugs per encounter’ and ‘percentage of encounters with an antibiotic’. These findings continued to highlight the importance of these indicators as a teaching tool for better practice.
Rational prescription
The importance of teaching and monitoring correct prescription techniques in developing countries where the tendency for multiple prescriptions is rampant cannot be overemphasized. Our experience working in developing countries where tendencies are: the more medication prescribed the better, injectables are superior to oral medication whether justified or not, and the more antibiotics the better, shows us the significance of the results we achieved with our programme.
The indicator ‘average number of drugs prescribed per encounter’ monitors the problem of multiple prescriptions. Everyone either reached or performed better than the norm except Mpona. While Mpona had respected national protocols in prescription we viewed them also as within the norm. Again, this emphasizes the use of these indicators as guidelines for teaching and not as fast and hard rules.
Initial data show a high percentage of generic drug prescription. This was mainly due to ICRC, as the principle provider of medication, donating only generic medication and the restricted access of health centre personnel to drugs on the open market. Interestingly however, when the Ikela front line reopened to free circulation (October), allowing access to private pharmacies, prescription of generic drugs remained 100% for five of the six health centres, indicating that they did not buy drugs on the open market. This signified their acceptance of the indicator.
Improvement and continued adherence to the indicator ‘percentage of encounters with an antibiotic’ demonstrates their improved knowledge regarding antibiotic use. Monkoso was the only centre remaining outside the norm; however they had a high prevalence of STDs with all prescriptions respecting national protocols.
Due to the numerous complications seen as a result of injections, and their tendency to overuse injectable medicine, we chose not to deliver injectable medicine to the health centres. However, as with generic drugs, the improving security situation allowed access to outside pharmacies, providing availability of injectable medication. Thus the improvement we saw in this category is indicative of their understanding and willingness to adhere to agreed-upon better practice. Although Anzi remained above the norm at the end of December (no epidemiological justification), throughout the year they were either at or below the norm.
Although three health centres remained above the no-tolerance norm with the indicator ‘percentage of encounters with a contraindication’, all health structures showed marked improvement. This improvement demonstrates not only their acceptance of the indicator but also their willingness to learn from the books and teaching sessions the ICRC team gave on rational prescription techniques.
Hygiene indicator category
From cleanliness of the latrines and surrounding area to hygiene of health personnel, this category was important in aiding staff to achieve and maintain good hygiene techniques. The marked improvement of five out of six health centres showed their willingness to accept and own these indicators. In the case of Bosango, continued monitoring of this indicator revealed poor maintenance of the existing structure and motivated the community to renovate the walls and roofs.
Use of these indicators enabled the hospital in Kalemie to pinpoint two problems with patient teaching and hygiene. Patients were throwing garbage out of the open windows from the in-patient wards. Many patients were also unfamiliar with indoor toilets and were unable to use them properly. With proper patient teaching by the nursing staff, slow improvement began.
Pharmacy management category
This category allowed us to gather data to assess the management of the pharmacy stock and follow consumption of each medication in order to avoid misuse or theft. This category remains one of the most important for the ICRC and health committee. The impressive improvement made by the centre and hospital throughout the year indicate their desire to follow this indicator of good practice. From early January 2003, they were trained and asked for monthly check-ups.
|
Limitations |
|---|
|
TopSetting and participantsDevelopment of methodologyAnalysisResultsDiscussionLimitationsImplicationsReferences |
|---|
The regular presence of an ICRC local doctor in the field to reinforce and monitor progress of the indicators was key to a successful beginning. Without the availability of a continued and regular presence, adherence and understanding of the implementation of the indicators may not have led to the results we found.
The main limitation for this study is whether the results would be replicable in other cultures. This would be especially interesting given our findings when we added the scoring system for indicator categories. The enthusiasm and fascination of the various health centre teams and hospital to know and improve their scores and to compare them with other health centres in their area provided strong motivation for improvement and came as a surprise to us. Whether other countries and cultures would be as enthusiastic and motivated to improve their scores is unknown. Indeed, in some areas, the real aim of the indicators had to be regularly re-emphasized as acquiring points became their goal rather than achieving better practice.
|
Implications |
|---|
|
TopSetting and participantsDevelopment of methodologyAnalysisResultsDiscussionLimitationsImplicationsReferences |
|---|
The positive effects of implementing quality improvement projects in a health structure are well known and documented. However, these effects are not known when translated to areas of conflict or in settings where staff reliability is affected by conflict or security incidents. Through creation, implementation, and acceptance of our quality improvement programme in the DRC, and with the documentation of our results, we hope to send encouraging signs to others that quality improvment projects bode well for future use in settings such as ours.
Involving the community through participation of health centre and hospital teams to design population-based indicators was a new approach. This involvement helped communities take ownership of the indicators, hence our findings that poor performance on indicators prompted communities to seek training and assistance to improve quality of care. Additionally, continued adherence and improvement in each category confirmed the long-term effects of teaching sessions in the areas of rational prescription, hygiene, and pharmacy maintenance.
At certain ICRC-supported health centres where implementation of this quality improvement programme has taken place, these indicator categories are now used to describe the level of quality achieved and point out areas where specific training is needed. The results are routinely presented at the monthly meeting between the health centre and the health committee.
By respecting the four pillars of quality assurance: priority to the clients, priority to the process, priority to decisions taken from data, and priority to team involvement [1], this method is proposed as a supervisory tool that will lead to improved quality of care, even in difficult or unstable conditions.
|
References |
|---|
|
TopSetting and participantsDevelopment of methodologyAnalysisResultsDiscussionLimitationsImplicationsReferences |
|---|
- Quality Assurance Project. QA Monograph: A Modern Paradigm for Improving Healthcare Quality. Bethesda, MD: Quality Assurance Project: http://www.qaproject.org Accessed 10 July 2004.
- Boonstra E, Lindbaek M, Khulumani P et al. Adherence to treatment guidelines in primary health care facilities in Botswana. Trop Med Int Health 2002; 7: 178–186.[Medline]
- Krause G, Borchert M, Benzler J et al. Rationality of drug prescriptions in rural health centres in Burkina Faso. Health Policy Plann 1999; 14: 291–298.
[Abstract/Free Full Text] - Phadke AR. The quality of prescribing in an Indian district. Natl Med J India 1996; 9: 60–65.[Medline]
- Van Damme WI, Van Lerberghe WI, Boelaert M. Primary health care vs. emergency medical assistance: a conceptual framework. Health Policy Plann 2002; 17: 49–60.
[Abstract/Free Full Text] - Lagoutte J. Quality of clinical activities in 15 dispensaries supported by ICRC in Angola (in French). ICRC report of mission 96/197. 1996 unpublished.
- Coninx R, Aerts A. Quantified evaluation in primary health care: the example from a war affected area in rural Mozambique. ICRC report of mission. 1996 unpublished.
- Perrin P. War and Public Health. Geneva: International Committee of the Red Cross, 1996.
- Banatvala N, Zwi AB. Public health and humanitarian interventions: developing the evidence base. Br Med J 2000; 321: 101–105.
[Free Full Text] - Palmer C.A. Rapid appraisal of needs in reproductive health care in southern Sudan: qualitative study. Br Med J 1999; 319: 743–748.
[Abstract/Free Full Text] - The Active Learning Network for Accountability and Performance in Humanitarian Action: http://www.alnap.org Accessed 10 December 2004.
- Goma Epidemiology. Group Public health impact of Rwandan refugee crisis: what happened in Goma, Zaire, in July, 1994? Lancet 1995; 345: 339–344.[CrossRef][Web of Science][Medline]
- Siddique AK, Salam A, Islam MS et al. Why treatment centres failed to prevent cholera deaths among Rwandan refugees in Goma, Zaire. Lancet 1995; 345: 359–361.[CrossRef][Web of Science][Medline]
- The Sphere Project. Humanitarian Charter and Minimum Standards in Disaster Response. Oxford: Oxfam Publishing, 2000. Also the web site: www.sphereproject.org Accessed 10 December 2004.
- Quality Assurance Project. Health Manager Guide: Monitoring the Quality of Primary Health Care. Bethesda, MD: Quality Assurance Project: http://www.qaproject.org Accessed 10 December 2004.
- World Health Organization. How to Investigate Drug Use in Health Facilities. WHO/DAP/93.1: http://www.who.int Accessed 10 December 2004.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Kotagal, P. Lee, C. Habiyakare, R. Dusabe, P. Kanama, H. M Epino, M. L Rich, and P. E Farmer Improving quality in resource poor settings: observational study from rural Rwanda BMJ, October 30, 2009; 339(oct30_1): b3488 - b3488. [Full Text]
|
|
|
|
 |
|
 E. Bradley, K. A. Hartwig, L. A. Rowe, E. J. Cherlin, J. Pashman, R. Wong, T. Dentry, W. E. Wood, and Y. Abebe Hospital quality improvement in Ethiopia: a partnership-mentoring model Int. J. Qual. Health Care, December 1, 2008; 20(6): 392 - 399. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||