A clinical practice i - OPUS at UTS - University of Technology Sydney [PDF]

Evaluation of a program implemented to reduce surgical wound infection in an acute care hospital in India: A clinical practice improvement project

Sr. Alphonsa Ancheril

A thesis submitted in accordance with the requirements for admission to the Degree of Doctor of Philosophy University of Technology, Sydney May 2004

CERTIFICATE OF AUTHORSHIP/ORIGINALITY

I hereby certify that this thesis entitled “Evaluation of a program implemented to reduce surgical wound infection in an acute care hospital in India: A clinical practice improvement project” is the outcome of the original research undertaken and carried out by me. I also certify that the material of this thesis has not formed in any way, the basis for the award of any previous Degree, Diploma, Title or Recognition.

I also certify that the thesis has been written by me. Any help that I have received in my research work and the preparation of the thesis itself has been acknowledged. In addition, I certify that all information sources and literature used are indicated in the thesis.

Signature of Candidate

………………………………………………….

ii

TABLE OF CONTENTS LIST OF TABLES ........................................................................................................vii LIST OF FIGURES .......................................................................................................ix LIST OF ABBREVIATIONS ......................................................................................x ACKNOWLEDGEMENTS.........................................................................................xi ABSTRACT ................................................................................................................. xiii CHAPTER 1: INTRODUCTION ...............................................................................1 Introduction......................................................................................................................1 Background of the study ..................................................................................................2 Health care in India .....................................................................................................2 Issues arising from nosocomial infections in India..........................................................5 Research Project.............................................................................................................10 Aims and Objectives...................................................................................................10 Research questions.....................................................................................................11 Setting of the study .....................................................................................................11 Significance of the research.......................................................................................12 Conceptual Framework ..................................................................................................12 Organisation of the thesis...............................................................................................16 Summary ........................................................................................................................17

CHAPTER 2: LITERATURE REVIEW ................................................................18 Introduction....................................................................................................................18 Part A: Nature of the clinical problem...........................................................................19 Summary ........................................................................................................................52 Part B: Selection of action research method ..................................................................54 Summary ........................................................................................................................61

CHAPTER 3: METHOD.............................................................................................62 Introduction....................................................................................................................62 Design of the study ........................................................................................................63 iii

Setting ............................................................................................................................66 Population and sample ...................................................................................................68 Instruments.....................................................................................................................69 Ethical considerations ....................................................................................................71 Diagnostic Phase (Phase 1)............................................................................................72 Procedure for collecting nursing data.......................................................................72 Procedure for collecting patient data ........................................................................72 Intervention Phase (Phase 2)..........................................................................................75 Evaluation Phase (Phase 3)............................................................................................76 Problems faced during data collection...........................................................................76 Overview of data analyses procedure ............................................................................77 Summary ........................................................................................................................78

CHAPTER 4: RESULTS OF DIAGNOSTIC PHASE........................................79 Introduction....................................................................................................................79 Category A: Patient variable..........................................................................................80 Category A: Surgical variables ......................................................................................81 Category A: Organisational variables............................................................................84 Category B Nurses variable ...........................................................................................85 Category C: Outcomes ...................................................................................................95 Category D: Surgical wound infection rate for each of the independent variables .......96 Summary ......................................................................................................................103

CHAPTER 5: INTERVENTION ............................................................................105 Introduction..................................................................................................................105 Into the field.................................................................................................................105 The intervention process ..............................................................................................107 Strategies to improve infection control........................................................................110 Sustainability of the program.......................................................................................116 Nurses responses to the intervention ...........................................................................118 Summary ......................................................................................................................119

iv

CHAPTER 6: COMPARISON OF PRE AND POST INTERVENTION RESULTS .....................................................................................................................120 Introduction..................................................................................................................120 Patient variables ...........................................................................................................121 Surgical variables.........................................................................................................122 Organisational variables...............................................................................................124 Nurse variables.............................................................................................................126 Content analysis of the observations and field notes ...................................................128 Category C: Outcome variable - Rates and severity of surgical wound infection.......131 Category D: Number and percentage of wound infection in group 1 and group 2 depending upon the various risk factors ......................................................................135 Category E: Prediction of a model that contributed to surgical wound infection........141 Summary ......................................................................................................................143

CHAPTER 7: DISCUSSION AND CONCLUSION.........................................144 Introduction..................................................................................................................144 Major findings of the study..........................................................................................145 Rate and severity of surgical wound infection.........................................................145 Risk factors that contributed to surgical wound infection.......................................146 Discussion of the findings and significant categories from the action research process..........................................................................................................................152 Implications of this study.............................................................................................158 Further research ...........................................................................................................167 Limitations of the study ...............................................................................................168 Conclusion ...................................................................................................................170

REFERENCES ............................................................................................................174 APPENDICES .............................................................................................................200 APPENDIX 1...............................................................................................................201 APPENDIX 2...............................................................................................................202 APPENDIX 3...............................................................................................................203 v

APPENDIX 4...............................................................................................................204 APPENDIX 5...............................................................................................................205 APPENDIX 6...............................................................................................................206 APPENDIX 7...............................................................................................................209 APPENDIX 8...............................................................................................................211 APPENDIX 9...............................................................................................................213 APPENDIX 10.............................................................................................................214 APPENDIX 11.............................................................................................................215 APPENDIX 12.............................................................................................................225 APPENDIX 13.............................................................................................................242 APPENDIX 14.............................................................................................................244 APPENDIX 15.............................................................................................................245 APPENDIX 16.............................................................................................................246 APPENDIX 17.............................................................................................................249 APPENDIX 18.............................................................................................................250

vi

LIST OF TABLES Table 1:

Number and percentage of patients operated by each surgeons

82

Table 2:

Number and percentage of patients according to wound class

82

Table 3:

Number and percentage of patients according to type of surgery

83

Table 4:

Number and percentage of patients according to duration of surgery

83

Table 5:

Number and percentage of patients according to time of surgery

83

Table 6:

Number and percentage of patients according to pre-operative stay days 84

Table 7:

Number and percentage of patients according to pre-operative shaving time

84

Table 8:

Distribution of nurses sample according to their age

85

Table 9:

Distribution of nurses sample according to their years of experience

85

Table 10: Hand washing score obtained by nurses

86

Table 11: Wound dressing score obtained by nurses

87

Table 12: Distribution of patient sample according to their post-operative stay

96

Table 13: Surgical wound infection rate according to gender and co-morbidity

96

Table 14: Surgical wound infection rate per surgeon

98

Table 15: Surgical wound infection rate and type of surgery

100

Table 16: Surgical wound infection rate and duration of surgery

100

Table 17: Surgical wound infection rate and time of surgery

101

Table 18: Predictors that contributed to surgical wound infection

103

Table 19: Distribution of group 1 and group 2 sample according to type of surgery 123 Table 20: Distribution of group 1 and group 2 sample according to duration of surgery

124

Table 21: Distribution of group 1 and group 2 sample according to time of surgery 124 Table 22: Distribution of group 1 and group 2 sample according to their preoperative stay

125

Table 23: Distribution of group 1 and group 2 sample according to pre-operative shaving time

125

vii

Table 24: Wilcoxon signed rank test to find out the difference in hand washing and wound dressing scores of nurses pre and post the intervention

127

Table 25: Comparison of group 1 and group 2 sample characteristics

132

Table 26: Difference in wound score pre and post the intervention

134

Table 27: Difference in post-operative stay days

135

Table 28: Surgical wound infection rate in group 1 group 2 according to gender

136

Table 29: Surgical wound infection rate in group 1 group 2 according to co-morbidity

137

Table 30: Surgical wound infection rate in group 1 group 2 according to type of surgery

138

Table 31: Surgical wound infection rate in group 1 group 2 according to duration of surgery

138

Table 32: Surgical wound infection rate in group 1 group 2 according to time of surgery

139

Table 33: Surgical wound infection rate in group 1 group 2 according to pre-operative shaving time

140

Table 34: Logistic regression analysis to predict significant variables contributing to surgical wound infection model

142

viii

LIST OF FIGURES Figure 1: Map of India

2

Figure 2: Conceptual framework

15

Figure 3: Relationship between quality improvement and quality assurance

45

Figure 4: Diagrammatic representation of research design

64

Figure 5: Map of Mangalore

67

Figure 6: Distribution of the patient sample by age category

81

Figure 7: Distribution of nurse’s sample by education

86

Figure 8: Content analysis of field notes and observations

88

Figure 9: Surgical wound infection rate and severity

95

Figure 10: Age of the patient and infection rate

97

Figure 11: Surgical wound infection rate and wound class

99

Figure 12: Surgical wound infection rate and pre-operative stay

101

Figure 13: Surgical wound infection rate and pre-operative shaving time

102

Figure 14: Distribution of group 1 and group 2 sample by age

121

Figure 15: Number of surgeries per surgeon in group 1 and group 2 sample

122

Figure 16: Distribution of group 1 and group 2 sample according to wound class

123

Figure 17: Hand washing score of nurses pre and post the intervention

126

Figure 18: Wound dressing score of nurses pre and post the intervention

127

Figure 19: Wound infection rate group 1 and group 2 sample

133

Figure 20: Severity of wound infection group 1 and group 2 sample

133

Figure 21: Surgical wound infection rate in group 1 and group 2 sample according to age category

136

Figure 22: Surgical wound infection rate in group 1 and group 2 sample according to wound class

137

Figure 23: Surgical wound infection rate in group 1 and group 2 sample according to pre-operative stay days

140

Figure 24: Diagrammatic representation of factors that contributed to surgical wound infection

143

ix

LIST OF ABBREVIATIONS APIC

Association for Practitioners in Infection Control

CDC

Centres for Disease Control and Prevention

CQI

Continuous Quality Improvement

EPINE

Nosocomial Infections Prevalence Study in Spain

HAI

Hospital Acquired Infection

HIC

Hospital Infection Control

ICPs

Infection Control Professionals

JCAHO

Joint Commission of Accreditation of Healthcare Organisations

MRSA

Methicillin Resistant Staphylococcus Aureus

NI

Nosocomial Infections

NNIS

National Nosocomial Infections Surveillance

NSW

New South Wales

PAR

Participatory Action Research

SENIC

Study on the Efficacy of Nosocomial Infection Control

SHEA

Society for Hospital epidemiology of America

SIS

Surgical Infection Society

x

ACKNOWLEDGEMENTS The researcher is extremely grateful to all those who have contributed to the successful completion of this project. My sincere and heartfelt thanks to Professor Judith Donoghue, Head of the department of Acute Care Nursing Research Unit, St. George Hospital, Kogarah, for her constant support, guidance, valuable suggestions, constructive criticism and encouragement given all throughout the project. She had been very patient with me and we have spent hours together to discuss, interpret the findings and refine this thesis. I am equally grateful to Professor Lesley Barclay, for her valuable guidance whenever needed and her sustained interest in the project as well as helping me to get a scholarship from the Faculty of Nursing, Midwifery and Health, University of Technology, Sydney (UTS). I am extremely grateful to the Faculty of Nursing, Midwifery and Health, UTS, not only for the professional advice and interest they showed in the project but also for providing me with a scholarship to undertake this study. I wish to express my love and gratitude to the staff of Acute Care Nursing Research Unit, St. George Hospital Kogarah, Family Health Research Unit, Broadway and my colleagues for their friendship and in particular Ms Priya Nair for her assistance with printing and binding of the thesis. A sincere thanks to Mr. Suresh, Statistician, for his expert guidance in statistical analysis and helping to workout the many statistical issues. A special thanks to my nephew Jinto, for helping me with computer typing and preparing the “booklet” and Mr. Christopher Sudhaker, my former student and friend, for helping in the preparation of the “Infection Control Manual”. Words are inadequate to express my sincere love and gratitude to Sr. Anne O’Shaughnessy and all the members of ‘Our Lady’s Nurses for the Poor’, who met all my needs while I undertook my study at UTS. They were not only kind enough to accommodate me but generously provided me with all the financial, moral, and spiritual xi

support I needed and made sure that I was not in want of anything. Thanks to all of you for your love, friendship, constant support and encouragement. I will be failing in my duty if I do not thank Sr. Patricia Davis for being a good friend of mine, constantly encouraging me and supporting me with her prayers and patiently editing my thesis. I am also grateful to the friends, relatives and well wishers of ‘Our Lady’s Nurses for the Poor’ for the sincere interest they showed in me and a special thanks to Diane for helping me with computer work as well as for the many little things she did to make my stay in Sydney a comfortable one. Thanks to Thomas & Lissy, Rojo & Asha, and many other Indian friends who are settled in Sydney, for their friendship, love, and support of me; each in different ways while I undertook my study at UTS. I owe my sincere thanks to Fr. Baptist Menenzes, Director of Fr. Muller Charitable Institutions and Fr. Lawrence C D’Souza, Administrator, Fr. Muller Medical College and College of Nursing, for deputing me for higher studies, for their constant help, support and encouragement and permitting me to undertake the study at the particular Institution. I wish to express my deep sense of gratitude to the Superior General, Provincial, Superiors and sisters of my religious family, my mother and dear ones, friends and well wishers, for their constant encouragement, help and prayerful support. A big thanks to the Nursing Superintendent and all the nurses who participated in the project. This project would not have been possible without their enthusiasm, co-operation and commitment. Finally my sincere thanks to the many people who directly or indirectly have contributed to my safe stay at Sydney, study at UTS and successful completion of this project.

xii

ABSTRACT This research project investigated the impact of an action research intervention implemented to reducing surgical wound infection in one of the acute care hospitals in India. The study aimed to develop and implement a clinical practice improvement program in reducing surgical wound infection by improving the hand washing and wound dressing practices of nurses. The study also aimed to identify the important contributing factors to a model that predicts surgical wound infection. Pre-post evaluation measures were taken to compare the results of surgical wound infection rate before and after the implementation of the intervention. Surgical wounds of two thousand patients (one thousand before the intervention and another one thousand after) were assessed to determine the wound infection rate and severity of wound infection. The hand washing and wound dressing practices of forty nurses were observed. These same nurses were involved in the intervention using a participatory action research process. The results of the study suggest that there was a marked, significant reduction in the rate and severity of wound infection following the implementation of the intervention. By increasing the hand washing facilities in the ward, educating nurses on the importance of better hygiene, pre-operative shaving and post-operative wound care, the hand washing and wound dressing practices of nurses improved considerably. These improvements resulted in a reduction in the number and severity of patients’ surgical wound infections. The study also examined the contribution of different factors to surgical wound infection in a Indian hospital. Significant predictive factors were the patients’ age, longer preoperative hospital stay, extended pre-operative shaving time before surgery, wound class, and co-morbidity of the patient. The identification of risk factors that contributed to increased surgical wound infection for example pre-operative skin preparation, preoperative hospital stay of the patient would help in taking appropriate measures at the ward level and organisation as a whole. Nosocomial infections extends to an unnecessary lengthy hospital stay, additional treatment increased mortality and morbidity, and increased cost to the patients and the nation as a whole. xiii

This project proved that educational mentoring, data surveillance processes and involving the nurses in an action research process were effective in enabling participants to improve their clinical practice and thereby reduce the incidence of patients’ surgical wound infections. Establishing infection control teams, ongoing surveillance and feedback to staff of nosocomial infection rates is an urgent need in all Indian hospitals. Organisational management, as a priority, need to provide funding and staff dedicated to undertaking this essential work. Health care professionals can no longer plead ignorance of a situation for which all have a moral and professional responsibility.

xiv

CHAPTER 1: INTRODUCTION Introduction Florence Nightingale, more than 100 years ago is believed to have said “No stronger condemnation of any hospital or ward could be pronounced than the single fact that zygotic (infectious) diseases has originated in it, or that such a disease has attacked other patients than those brought in with them” (Shriniwas, 1992: page 1). No hospital should continue to allow this damage to occur to its patients. However hospital acquired infection is currently a worldwide problem, growing in impact rather than decreasing with modern health care. Nosocomial Infections (NI) otherwise known as Hospital Acquired Infections (HAI) are infections that are not present or incubating when the person is admitted to a health care facility (WHO 1988). The four main types are bacteraemia (intravenous line infection), catheter related urinary tract infection, respiratory tract infection and surgical wound infection. Surgical wound infection accounts for one fifth of all nosocomial infections. According to Ganguly et al., (2000) at any point of time more than 1.4 million people globally suffer from nosocomial infections. This figure represents only the tip of the iceberg as the record keeping systems in most developing and / underdeveloped countries are poorly evolved and managed (Ganguly et al., 2000). Nosocomial infections occur in about 5% to 10% of hospital admissions worldwide. In India, the nosocomial infection rate is alarming and is estimated to be about 30% to 35% of all hospital admissions (Mukherjee, 2001). Again, this estimate may be less than the real level because of the lack of systematic evaluation. A study conducted by Ganguly et al., (2000) in one of the hospitals in Utter Pradesh, India, showed that out of 422 surgical patients 164 developed nosocomial infection, an overall incidence of 38.8%. These results illustrate that surgical wound infection is a major problem in Indian health care institutions where wound infection rates have been identified. If research to calculate national rates of surgical wound infection were undertaken, it is likely that similar estimates would be found in other hospitals irrespective of the State or the type of hospital. Unfortunately, rates that have been identified in India between 27% and 39% are higher when compared with those from Chapter 1 Introduction

1

other countries where surgical wound infection rate is around 5-10% (Kandula & Wenzel, 1993; Leu, 1995; McLaws et al., 1997; Murphy & McLaws, 1999; Astangneau, et al., 2000). Background of the study Health care in India India, the second most populous country in the world, has no more than 2.5% of the global land (3,287,263 sq. km) but is home to one sixth of the world's population.

Figure 1. Map of India

According to available information, India's population is over one billion. The population density is 274 persons per square kilometre. The demographic profile of India shows a young population (about 40% of the population is below the age of 14 years), a low Chapter 1 Introduction

2

literacy status of 42%, declining mortality rate (9.5 per 1000 population), and dependency ratio of 0.9. These data indicate that every economically productive member has to support almost one dependent person (Sample Registration System, Government of India, 2000). According to Gupta (2003), India’s public expenditure on health is one of the lowest in the world. The government spends 17.3 per cent of its total health budget on public health system. By comparison England spends 96.9 percent, the United States allocates 44.1 percent, Sri-Lanka earmarks 45.4 percent and China’s 24.9 percent. According to World Health Organisation, only 0.49 percent of the population in India has access to essential drugs, whereas Pakistan and Bangladesh, known for their poor performance in the health sector, have a much higher rate with 50 to 79 percent of their population having access to essential drugs (Bal, 2003). During the last few decades there has been notable improvement in the health status of the Indian population. The death rate has steadily declined from 11.7 in 1985 to 9.5 in 1993. The life expectancy at birth has gone up considerably from 56 years to 65 years. The mortality rate for a number of communicable diseases such as cholera, malaria, plague have also declined. However India's health standards are still low compared to those in developed countries. The infant mortality rate (IMR) in developed countries is as low as 4 per 1000 live births per year. In India it continues to be high at 71 per 1000 live births per year. However there are wide interstate variations in India in rates such as the IMR. For instance in the State of Uttar Pradesh the IMR is 97 whereas in Kerala State it is 12 (Ninth Five Year Plan of India, 1997). In spite of the impressive gains in health care status reflected by the statistics, the demographic and health picture of the country is still a cause for serious and urgent concern. Various communicable and non-communicable diseases have to be brought under control. The morbidity is dominated by communicable diseases such as tuberculosis, malaria, human immuno-deficency virus, viral hepatitis B, acute respiratory infections, diarrhoea, measles, neonatal tetanus, pertussis, filaria, leprosy, and Japanese encephalitis. Malnutrition, vitamin A deficiency, Iron-deficiency anaemia and IodineChapter 1 Introduction

3

deficiency disorders are also of great concern. Other common diseases are cholera, dengue, and rabies. Non-communicable diseases such as cancer, cardiovascular diseases, diabetes and hospital acquired infections are on the increase. Several national programs are run directly by the Ministry of Health to reduce the mortality and morbidity caused by such health problems (Planning Commission of India, 1996). A major requirement in Indian health care is developing adequate infrastructure and health personnel. According to data available in 1993 from the Directorate of Health Service, there were 13,692 hospitals and 596,220 hospital beds in India (Planning Commission reports, 1996). Current problems faced by the health care services include: •

persistent gaps in person power and infrastructure at the primary health care level;

•

suboptimal functioning of the infrastructure, poor referral services;

•

a plethora of Government, voluntary and private sector hospitals not having appropriate person power, diagnostic and therapeutic services and drugs;

•

massive inter-state / inter-district differences in performance as assessed by health and demographic indices: availability and utilisation of services are poorest in the most needy states / districts;

•

sub optimal inter-sectoral coordination;

•

increasing dual disease burden of communicable and non-communicable diseases because of ongoing demographic, lifestyle and environmental transitions;

•

technological advances which widen the spectrum of possible interventions;

•

increasing awareness and expectations of the population regarding health care services;

•

escalating costs of health care, widening gaps between what is possible and what the individual or the country can afford (Planning Commission of India, 1996).

The Ninth Five Year Plan of India (1997-2002) suggested many approaches to improve the health status of the population and the quality of care provided to patients. Some of the suggestions made were to: •

improve the efficiency of existing health care infrastructure at the primary, secondary, and tertiary care setting, through appropriate institutional strengthening, improvement of referral linkages and reliable Health Management Information System;

Chapter 1 Introduction

4

•

promote the development of human resources for health that are adequate in quantity, appropriate in quality, and accessible to all, so that there is improvement in the health status of the communities;

•

organise programs for continuing education in health services, update knowledge and upgrade skills of all workers and promote cohesive team work;

•

improve effectiveness of existing programs for the control of communicable diseases to achieve horizontal integration of ongoing vertical programs at the district and below district level;

•

strengthen the disease surveillance with focus on rapid recognition, reporting and response at district level;

•

promote production and distribution of appropriate vaccines of assured quality at affordable cost;

•

improve water quality and environmental sanitation;

•

1

•

enhance research capability with a view to strengthening basic, clinical and health

improve hospital infection control and waste management;

systems, research aimed at improving the quality and outreach of services at various levels of health care (Planning Commission of India, 1996) The Ninth Five Year Plan of India (1997-2002) has suggested improved nosocomial infection controls, as one of the many approaches to improve the quality of care provided to patients. However, no systematic programs or organisational processes have been instigated on either a State or National basis. There is no available data therefore to show the severity of the problem of NI throughout Indian hospitals. Issues arising from nosocomial infections in India Nosocomial infection is a major international public health problem. A WHO sponsored international survey in 1985-86, involving 14 countries across five continents, indicated that the prevalence rate of nosocomial infection varied from 3% to 21% with a mean of 9% among patients admitted to hospitals. There are no national data estimates of prevalence rates in India as there are no mandatory reporting and recording systems for

1

The underlining was added by the author to emphasise the relevance to this research thesis

Chapter 1 Introduction

5

nosocomial infections. However, there is little doubt that the incidence of nosocomial infections has increased over the years and the problem is no less in this country than anywhere else in the world. Moreover, there are likely to be wide variations from hospital to hospital, from one unit to another within a single hospital, and within the same unit over a period of time. Since policies for control measures depend on the prevalence of specific organisms in a particular hospital, it is imperative that hospitals run ongoing nosocomial infections surveillance programs to collect such data (Nanivadekar, 1997. The need to establish such recording process is imperative, as it is well established from international research that nosocomial infections results in extended and unexpected hospital stays, added burdens of pain and suffering to the patient as well as relatives, increased cost for the individual/hospital, and deprivation of treatment for deserving patients (Mukherjee, 2001). The strong religious background of the people in India make them believe that what happens to them is their fate and they should not blame anyone for it. This, coupled with trust in the health team and lack of knowledge of legal systems, contribute to the low level litigation subsequent to nosocomial infections. Laboratory tests are costly and many Indian hospitals lack the infrastructure to conduct most of these tests in their own hospitals. Consequently, doctors rely on clinical findings and the wide use of antibiotics. There is no national health insurance scheme in India and most patients pay for their health expenses. If a patient needs expensive antibiotics or has a prolonged hospital stay due to a hospital acquired infection, the cost is borne normally by the patient rather than the hospital (except for the poor, in which case the hospital subsidises the costs). Nosocomial infections have been globally recognised for over a century as a critical problem affecting the quality of health care and a principal source of adverse health care outcomes. Among all major complications of hospitalisation, nosocomial infections account for 50% (Gaynes, 1997. The third leading cause of nosocomial infection is surgical site infection which is estimated as 24% of all nosocomial infections (Pittet & Ducel, 1994; Mangram et al., 1999). Twenty four percent is an exceedingly high incidence, as a level of no greater than 5% is considered an acceptable incidence of nosocomial infection. The prevalence and risk factors of surgical site infections in many Chapter 1 Introduction

6

developing countries such as India have not been determined (Nanivadekar, 1997). According to Gupta (2003) a very large number of Indians are dying every year due to nosocomial infections. The patient is not aware that this disease has resulted from hospitalisation and suffers for no fault of their own. The main causes postulated for the development and spread of nosocomial infections in India are the indiscriminate use of antibiotics, failure of many hospitals to follow basic infection control methods like washing of hands, overall lack of hygiene in public hospitals and the poor State of government hospitals (Mukherjee, 2001). India has neither stringent laws on the sale of antimicrobials, nor notification systems, which in infections like Methicillin Resistant Staphylococcus Aureus (MRSA), are necessary. A retrospective study conducted in one of the cancer hospital in Bombay, India revealed a sudden climb of MRSA in surgical patients from 35% in 1988 to 64% in 1996 (Kelkar, 1997). Unfortunately, comparative data are not available even from other major Indian hospitals, because the majority lack the infrastructure and/or resources for monitoring nosocomial pathogens and maintaining infection control programs. The problem of nosocomial infection has received the attention of the Indian Government for decades. Two influential, national committees, one in 1968 headed by Dr. K.N. Rao, the then Director General of Health Services and other in 1976 headed by Dr. Sharad Kumar, Deputy Director of Health Services, investigated in detail the problems of hospital infection in Delhi hospitals. Dr. Rao’s committee emphasized that “The reservoirs of infections in the hospitals must be attacked, carriers must be dealt with, and rigorous asepsis in the wards and theatres introduced. Clean air, clean beddings and hygienic methods of dust removal must be recognised as the basic requirements, and the use of hospitals must be strictly controlled and dictated by essential needs” (Shriniwas, 1992: p.2). In 1976, the Sharad Kumar committee recommended the formulation of Hospital Infection Control Committees, maintenance of proper medical records, and medical audit (mortality committee), training of staff, control of overcrowding, improvement of sanitation, kitchen and laundry services (Shriniwas, 1992). These recommendations have largely remained unaddressed in many States.

Chapter 1 Introduction

7

The Planning Commission of the Indian Government identified nosocomial infection control as an issue to be dealt with during the Government's Ninth Five Year Plan (19972002). But no programs or guidelines are laid down at the National or State level to deal with these issues. Few Indian hospitals have an Infection Control officer or any nosocomial infection control programs (personal communication with the Health Department, Mangalore, 2001). The small number of hospitals that have implemented nosocomial infection control program have found it to be very effective in reducing hospital acquired infections (Kelkar, 1997; Nanivadekar, 1997; Ganguly et al., 2000). Although all nosocomial infections should be considered preventable, complete prevention remains unrealistic in practice. Emphasis has been laid on simple measures like hand washing and education of health care workers, ongoing training and surveillance systems and utilisation of the services of an infection control person. If adequate measures are taken the occurrence and severity of infection can be reduced significantly. Data available from other developed and developing countries demonstrate that nosocomial infections can be controlled and prevented if adequate measures are taken (Pittet et al., 1995; Siddiqui & Luby, 1997; Huskins et al., 1998; Finkelstein et al., 2000; Kim et al., 2000). In spite of the fact that the importance of prevention and control of nosocomial infections has been recognised at the highest level, the actual incidence of nosocomial infection in hospitals of India is not known. While there is a paucity of literature to demonstrate the seriousness of the problem in Indian hospitals, there are varying opinions among health personnel regarding nosocomial infections and the recognition of the seriousness of the problem. For instance in an Indian newspaper (The Tribune, July 11, 2001), Dr. J. D. Wig wrote that hospital stay can make one 'more sick'. Hospitalisation in India frequently adds distress and risk to patients when health care is not as good as patients expect and hope it to be. Under Indian conditions more and more patients admitted to hospital acquire infections from those hospitals. The mushrooming of small hospitals and nursing homes in the country, with inadequate facilities also contributes to the increasing incidence in hospital-acquired infections (The Tribune, India-Health Tribune, 2001).

Chapter 1 Introduction

8

There are health professionals who fail to consider the seriousness of the problem or acknowledge that such a problem exists in their employing hospital. An anecdote confirms this situation. When the researcher discussed the problem of nosocomial infection and sought permission to conduct a study in a particular hospital in India, the first reaction of one of the senior health personnel was to ask why the researcher wanted to do this study in the hospital. The enquirer asked whether the researcher thought there was any problem with regard to patient care and was surprised to consider this might be a problem. When a pilot study on surgical wound infection was conducted in the same facility, and the results were presented to the same person, the response was one of amazement. This manager never imagined this particular hospital, which is well known for its high standards of patient care, had such a high rate of surgical wound infection (27.6%). The manager then requested the researcher to select this hospital for the study and do something to control and prevent surgical wound infection This is what she wrote to the researcher: “I thought why this study in our hospital. We may not have many infections. But when I saw the findings I got a shock (Letter 26 November 2000). Research into the reduction of nosocomial infections needs to consider that: y

nosocomial infection requires a source of micro-organisms,

y

a new host and

y

a chain of transmission (Meakins, 1994).

Nurses and other health care staff need to ensure that they are not part of that chain. Studies by various researchers, for example as cited by Pursell (1996), have found that the practices of some health professionals put patients at risk from infection, through both a lack of attention to cleanliness of equipment and poor hand-washing. Goldmann et al., (1992) found that the hands of 68% of nurses in a neonatal intensive care unit had heavy gram-negative bacterial colonization with the same strains as the babies for whom they were caring. One consistent theme that runs through the research findings is that hospital staff can be a major source of nosocomial infection, and infections can be reduced if staff take notice of research evidence and implement the findings (Pursell, 1996). It is therefore important that more hospitals take up studies on nosocomial infection to Chapter 1 Introduction

9

promote a better understanding of the local issues and provide a basis from which to develop effective interventions within their own health care system. The small pilot study referred to above, was conducted over a period of six months in an acute care hospital in Karnataka, India. The results showed a surgical wound infection rate of 27.6% (Unpublished data, 2001). This rate is very high compared to an overall infection rate of 5% which is considered to be acceptable (Nanivadekar, 1997: Pittet & Ducel, 1994) and confirmed the importance of undertaking the study reported in this thesis. The paucity of Indian literature on nosocomial infection, the high rate of surgical wound infection identified by the pilot study and the interest shown by some health personnel encouraged the researcher to take up the study. The study aimed to find out the surgical wound infection rate and to implement a program to reduce the incidence of surgical wound infection. The study was conducted in one of the large acute care hospitals in Mangalore. Research Project The topic for the project was formulated as “Evaluation of a system implemented to reduce surgical wound infection in an acute care hospital in India: A Clinical Practice Improvement Project”. Aims and Objectives The aim of the study, to develop and implement a practice improvement process to reduce surgical wound infection, was achieved through improving the hand washing and wound dressing techniques used by nurses. The effectiveness of the program was evaluated by measuring the rate and severity of surgical wound infection before and after the program was implemented. The study also aimed to identify the important contributing factors of a model predicting surgical wound infection.

Chapter 1 Introduction

10

Objectives of the study are: 1. To effect a change in the professional behaviours of nurses by improving the hand washing and wound dressing practices, to improve the outcomes for surgical patients 2. To record the rate and severity of surgical wound infection before and after a practice improvement intervention in a large Indian hospital 3. To identify any significant associations between surgical wound infection and factors identified from the literature such as, patient related factors (gender, age, comorbidity), surgery related factors (wound class, emergency versus elective surgery, duration of surgery, time of surgery), and organisational factors (pre-operative stay days, pre-operative shaving) to develop a predictive model for local use 4. To identify the theoretical categories of the intervention process to direct ongoing and sustainable improvement Research questions The research questions were: 1. Did an intervention to improve the nurses’ hand washing and surgical wound dressing practices significantly change their behaviours? 2. Are there significant differences in the infection rate and severity of surgical wound infection following the introduction of the intervention? 3. Are there any patient, surgical or organisational variables that predicted the severity of surgical wound infection? 4. What are the major theoretical categories identified from the action research process of the intervention? Setting of the study One of the acute care hospitals in Mangalore, India was selected for the study. Mangalore is one of the districts in Karnataka State, located in south India on the west coast. The population of Mangalore is around 759,705 (1991 Census). There are eight large hospitals (beds >500), seven medium sized (beds 250-500) and 17 small (beds 500 beds (5 large size hospitals). Significantly higher nosocomial infection prevalence rates were noted in larger hospitals and medium size (11.3%) compared to small hospitals (7.6%). The overall prevalence rate was 10.4% (Sax, 2000). The infection rate in developing countries / underdeveloped countries continues to remain as high as 25% to 35% (Neuber, 1996; Huskins et al., 1998). Hospitals in Mexico have an infection rate ranging between 15.8% to 31.3% (Ponce, 1991). However there are countries that have reported a low infection rate. For example one hospital in Thailand has reported an infection rate of 4.8% to 5.8% (Leu 1995). Nosocomial infection rate in one of the hospitals in Turkey was found to be 2.5%. This low rate of infection compared to other Turkish hospitals may be due to the fact that the study was conducted in a new medical

centre

with

limited

Chapter 2 Literature Review

bed

capacity,

and

organ

transplantation

and 20

immunosuppressive treatments were infrequent in this hospital (Durmaz et al., 2000) while in West Indies it was found to be 11.0% to 14.9% (Orrett et al., 1998). These are only isolated instances and representative of developing countries. In India no national figures are available to demonstrate the seriousness of the problem, however the rate is believed to be around 35% by some commentators (Ganguly et al., 2000; Wig, 2001). Nosocomial infections result in substantial morbidity, mortality, and cost. The excess duration of hospitalisation, secondary to nosocomial infections, has been estimated to be one to four days for urinary tract infections, seven to 8.2 days for surgical site infections, seven to 21 days for blood stream infections, and 6.8 to 30 days for pneumonia. The estimated mortalities associated with nosocomial infections are 14.8% to 71%. The cost to treat each nosocomial infection ranges from US $583 to US $4,886 (Jarvis, 1996). According to a recent report from UK, patients who developed a nosocomial infection remained in hospital two and a half times longer and incurred hospital costs almost three times higher than patients who didn't develop a nosocomial infection (Plowman, 2000; Plowman et al., 2001). These findings support the findings of other researchers, that nosocomial infections cause substantial morbidity, prolonged hospital stay, increased mortality, and cost (Rello, 1999; Raymond & Aujard, 2000; Mylotte et al., 2001; Astagneau et al., 2001). Rigorous studies of the impact of hospital infections on mortality and hospital costs in low middle income (LMI) countries have not been performed, but available estimates suggest these infections are a substantial public health problem. Ponce (1991) reported that 32 deaths per 100,000 inhabitants in Mexico are attributable to hospital acquired infections, potentially ranking these infections as the third most common cause of death. Based on an average prolonged length of stay of 10 days and an average bed-day cost of $100, he estimated that the yearly cost of hospital infections in Mexico approached $450 million excluding the cost of antimicrobial therapy (Huskins et al., 1998). As developing countries introduce the latest surgical techniques and a range of medical technology, often without an infection control infrastructure, more nosocomial infections will occur. National health authorities and hospital administrative personnel need to provide assurance that sufficient personnel and resources are provided for infection Chapter 2 Literature Review

21

control and prevention activities if increased morbidity, mortality, and cost expenditures associated with these infections are to be averted (Jarvis, 1996). b) Surgical wound infection Surgical wound infections are always nosocomial and are the third most frequently reported nosocomial infection (Davidson, 1995; Mangram et al., 1999). Prior to the late nineteenth century infection was common with all surgical procedures and lifethreatening infection was expected with any major operation. It was Joseph Lister's significant contribution to the understanding of infection that paved the way for safer surgery. Using Pasteur’s germ theory, Lister hypothesised that invisible bacteria in the environment could cause surgical wound infection if they gained access to tissues through broken skin. Thus, micro-organisms must be prevented from entering a wound during or after surgery. Further, micro-organisms already present in a wound at the time of surgery must be prevented from spreading (Meakins, 1994). Despite Lister's work and many subsequent advances, such as improvements in pre-operative preparation of the patient, anaesthesia and surgical technique, and the use of antibiotics for prophylaxis, surgical site infection remains a serious problem leading to many adverse effects. First of all, it leads to local physical complications such as delayed healing, tissue destruction, wound dehiscence, incisional and deep hernias, deformity, and pain (Meakins, 1994). Secondly, systemic complications like toxaemia, bacteraemia, septic shock, vital organ failure and death can occur as a result of surgical site infection (Meakins, 1994; Massnari et al., 1995; Astangneau et al., 2000). Thirdly, it can give rise to impaired psychological functioning leading to impaired quality of life (Meakins, 1994). Finally, economic problems as a result of loss of the productivity, prolonged hospitalisation, and increased health care costs are very high (Meakins, 1994; Astangneau et al., 2000). However, the actual incidences vary from country to country, hospital to hospital, surgeon to surgeon, from one surgical procedure to another because the criteria / definitions used to define surgical site infection may vary between and even within countries, and most importantly from one patient to another (Holzheimer et al., 1997). Researchers have defined surgical site infections differently. For instance Lujungqvist, defined surgical wound infection as a clear collection of pus, which empties itself Chapter 2 Literature Review

22

spontaneously or after incision (Byrne et al., 1988). This definition is inadequate as minor infections are difficult to classify satisfactorily. Based on certain criteria the CDC (Horan, et al., 1992; 1993; 1997; Mangram et al., 1999) has classified surgical wound infection as superficial incisional, deep incisional and organ / space surgical site infection. The detailed Table for defining surgical wound infection based on CDC is presented in Appendix 7. Wilson et al., (1988) defined surgical wound infection as wound infection with separation of deep tissue, and / or erythema, and / or purulent discharge, and / or serous discharge, and / or isolation of bacteria, and / or in-patient stay of more than 14 days. Scores are given for each of these criteria. Scores were also awarded to any additional treatment (antibiotics, drainage of pus under local anaesthesia, debridement of wound under general anaesthesia). Based on the total score, wound infection is classified as mild, moderate, or severe. The measuring instrument was named ASEPSIS, each letter representing one of the criteria. The advantages of the ASEPSIS instrument is that it has objectivity, reproducibility and its grading of wound infection helps to determine the severity of wound infection, instead of recording just the presence or absence of wound infection. This tool is used for the research presented here and is explained in more detail in the methodology section of this thesis. It can be seen from the above definitions that defining surgical site infections is complex. Criteria selected to define infection should be understood and followed by all staff in the particular hospital, unless a national standard is set by the government. Ideally, in each hospital the definition should remain unchanged, so that epidemiology and surgical staff can compare data over time and can evaluate interventions implemented to reduce rates. The definition should be simple to use and accepted by nurses and surgeons. Reported incidence of surgical wound infection A surgical wound infection rate of 11.2% was reported from one of the hospitals in Japan and 9.6% in another hospital (Morikane et al., 2000). Though reduction in surgical wound infection was negligible (10.1% from 11.2%) after the introduction of surveillance programs, it was noted that the severity of infection was reduced to a great extent, and most were only superficial incisional infections. A prospective survey of post-operative Chapter 2 Literature Review

23

nosocomial infections in one of the hospitals in Italy (1995) showed an incidence rate of 21.3 for 100 people discharged (Marena et al., 1997). The findings suggested the need to follow strict aseptic measures and to improve surveillance activities. Further hospitalwide surveillance of nosocomial infections following CDC definitions showed an incidence rate of 3.4% nosocomial infections. The main prevalent NI site was surgical site infection (0.83%) (Marena et al., 1998). These sudden decreases in rate may be due to the surveillance activities (Hawthrone effect) undertaken by the hospitals. DelgadoRodriguez et al., (1997) found that surgical infection rate in one of the hospitals in Southern Spain was 10.5%. A survey in Netherlands acute care hospitals (1996-1997) showed that out of 18,063 patients who had been surveyed, 562 patients developed surgical wound infection (Geubbles et al., 2000.). A community hospital in Breda, Netherlands found that of the 272 operated patients 18 developed surgical site infection (Kalmeijer et al., 2000). A five year period study (1985 - 1989) conducted at the University Hospital of Iowa showed that the infection rate after abdominal surgery was 10.5 per 100 procedures (Kandula & Wenzel, 1993). Comparison of infection rates in the European countries showed that Germany had the lowest rate (3.5%) of infection and Belgium had the highest rate (9.3%) of infections (Gastmeier et al., 1998). Nosocomial infections occurred in 2% to 5% of hospitalised patients in developed countries. These rates are considerably higher in developing countries, often reaching as high as 25%. According to Pittet & Ducel, (1994) and Nanivadekar, (1997) the overall incidence of surgical wound infection should not exceed more than 5% and in clean wounds 1.5%. In developing countries the magnitude of the problems have yet to be determined and many countries remain largely ignored. Velasco et al., (1996) conducted a study in a Brazilian tertiary cancer care hospital. The study aimed to develop a risk model for predicting surgical site infections in cancer patients undergoing operative procedures. Over a 24-month period, 1205 patients underwent operations for malignant disease. The overall surgical site infection rate was 17.3%. A prevalence survey of surgical wound infection in one of the hospitals in Vietnam showed that the surgical wound infection rate was 15.4% (Sohn et al., 2000). Chapter 2 Literature Review

24

The National Cancer Hospital in Mexico observed a surgical site infection rate of 9.30 per 100 operations (Vilar-Compte et al., 2000). A high rate of surgical wound infection (15%) following orthopaedic surgery was reported from a hospital in Pakistan in 1993. Following implementation of a surveillance program in 1994 the rate fell to 6% (Siddiqui & Luby, 1997). This situation is likely to be similar in India. International and national hospital surveys conducted in low middle income (LMI) countries demonstrate that hospital infection is one of the most common adverse outcomes of hospitalisation, occurring in 6% to 17% of patients (Huskins et al., 1998). At an International Conference in London (1986), Mayon-White from WHO Geneva, reported on a survey of 28,861 patients in 47 hospitals from four developing countries. Infection prevalence rates in individual hospitals ranged from 3% to 21%, with a mean of 8.7%. Prevalence rate of post-operative infection in individual hospitals ranged from 5.2% to 34.4% with a mean of 16.6% (HIC, 1987). This rate is very high compared to developed countries which has reported surgical wound infection rate of 3.5% to 9.3%. Developing countries need to establish surveillance systems to prevent and control surgical wound infection. Post discharge surgical wound infection Surveillance of surgical wound infection after patient discharge is also important as more procedures are performed on an out-patient basis and hospital stays grow shorter. Approximately one third of surgical wound infections occur after hospital discharge (HIC, 1991). A study conducted in Brazil from 1988 to 1992 showed that 32.2% to 91.4% who came to the out-patient clinic had developed surgical wound infection after discharge from the hospital (Ferraz et al., 1995). Similar findings were also were observed by other researchers (Zoutman et al., 1990; Gravel-Tropper et al., 1995; Taylor et al., 1998; Fields, 1999). These high rates are not acceptable and health care managers need to be concerned with such outcomes. Gravel-Tropper et al., (1995) found that with increasing number of same-day surgery and shortened hospital stays, it becomes more likely that a significant percentage of surgical wound infections will occur after patients are discharged from the hospital. They studied 469 patients who underwent surgical procedures at the Ottawa General hospital, in Chapter 2 Literature Review

25

Canada and 24 (5.2%) infections were detected. Of these, 14 (58.3%) were detected by the usual surveillance and 10 (41.7%) were detected by a physician after patients’ discharge. Failure to include post-discharge surveillance incidences results in a substantial underestimation of the true surgical wound infection rate. A Brazilian study from 1988 to 1992 showed that patients who came to the Out Patient clinic with surgical wound infection 32.2% to 91.4% developed wound infection after discharge from the hospital. In most patients (87.6%), surgical wound infection was detected within 15 days after surgery. The CDC have suggested that discharged patients should be contacted 30 days after operation to assess the presence or absence of infection (Ferraz et al., 1995). A post-discharge survey was conducted by Taylor et al., (1998) in Edmonton, Canada, to determine the frequency of surgical site infection after discharge from the hospital. Surgeons were contacted to find out if patients after discharge from hospital, had developed surgical wound infection. A total of 763 cases were surveyed and of this 55 (7.2%) had developed wound infection after discharge from the hospital. Therefore it can be argued that surgical wound infections are at a higher incidence than what is generally believed. In India there are no consolidated figures to assess the size of the problem. There is no reporting system for nosocomial infections (Nanivadekar, 1997). However, a study conducted by Ganguly et al., (2000) in India showed that out of 422 surgical patients 164 developed nosocomial infection. Most of the infections acquired by patients were in surgical wounds, in the respiratory tract or urinary tract. The overall incidence of nosocomial infection was 38.8% Overall reported rates of surgical site infections from developed countries range from 3.7% to 21.4%. This second figure represents a very high rate (Marena et al., 1997). But in developing countries the rate is unknown as many countries do not have a system of reporting nosocomial infections. There are a few studies available that show an infection rate as high as 38.8% (see for example HIC, 1987; Ganguly et al., 2000). Therefore it is important to identify and implement systems that will allow ongoing surveillance and

Chapter 2 Literature Review

26

bench marking internationally, in order to achieve best practice outcomes in Indian hospitals. c) Consequences of surgical wound infection Various studies have proved that nosocomial infections cause substantial morbidity, prolonged hospital stay, increased mortality and cost (Salemi et al., 1995; Poulsen et al., 1995; Raymond & Aujard, 2000; Astangneau et al., 2001). Patients who develop surgical site infections have a longer and costlier hospitalisation than patients who do not develop such infections. They are twice as likely to die, 60% are more likely to spend time in an ICU, and more than five times likely than the average surgical patient to be readmitted to the hospital (Kirkland et al., 1999 ). According to a recently reported study from the UK, patients who developed a nosocomial infection remained in hospital two and half times longer, incurred hospital costs almost three times higher, and incurred higher general practitioner, district nurse, and hospital costs after discharge from hospital than uninfected patients (Plowman, 2000). Patients with surgical wound infection remained in hospital three and half times longer than those who did not have an infection (Kandula & Wenzel, 1993; Merle et al., 2000; Geubbels et al., 2000). According to Zoutman et al., (1998) patients with surgical wound infection had an extra 10.2 days hospital stay and the costs attributable to wound infection was US $3,937 per infection. The Dutch study found an attributable increased length of stay of 8.2 days for patients with SSIs (Gaynes, 2000). The mean extra-hospital stay attributable to SSI was 8.2 days and the crude mortality rate was 1.4% (Ferraz et al., 1992; Barry & David, 1996; Jarvis, 1996; Marin et al., 1997; Severijnen et al., 1997; Astangneau et al., 2000; Mylotte et al., 2001). Experts in America estimate the cost for treating open surgical wound infection following abdominal surgery in America to be around US $2000 per patient. On an average in healthy non-obese patients, 2.5% of surgical wounds get infected. Therefore, obese patients would have a higher rate as obesity is found to be one of the risk factors for increased infection rate. The lower rate of infection costs a staggering cost of US $320,000 per year in the United States alone to treat infected open abdominal surgery incisions. The New York hospital study reported that adverse events complicated 3.7% of Chapter 2 Literature Review

27

hospitalisation and 16% of these events caused permanent disability or mortality. Massanari et al., (1995) maintain that at least 27% of the adverse events could have been prevented According to Gaynes (1997) in USA there are 90,000 deaths per year from hospital infections in the population of 32,000,000 hospital in-patients. The death rate from hospital acquired infection is 2.8 deaths per 1000 while death rate from heart disease is 2.7 per 1000 (Gaynes, 1997). The mortality rate related to nosocomial infection was 6.5% in one of the hospitals in Italy (Marena et al., 1998). The above mentioned studies clearly show that surgical wound infection has an impact on the patients, organisation and nation as a whole. In India most of the cost for treatment is met by the patients and their family. Extra hospital stay, treatment, morbidity and mortality due to hospital acquired infection adds extra burden to the family. Therefore, hospital acquired infection is an important area which needs attention in Indian hospitals. d) Risk factors contributing to surgical wound infection A number of risk factors for surgical site infection are identified in the literature. The factors are generally categorised as either host related, operation related, health personnel related or organisational factors Host related factors Age is considered to be one of the factors that increases the chances for surgical wound infection Aged people and very young children are more prone for infection (Hopfer, 1988; Mishriki et al., 1990; Nichols, 1992; Mclean & Dixon, 1994; Mertens et al., 1994; Danchaivijitr et al., 1995; Moro et al., 1996; Fernandez et al., 1997; Taylor et al., 1998; Bertin et al., 1998; Kirkland et al., 1999; de Boer et al., 1999; Gaynes, 2000; Chalfine et al., 2000). The weight of evidence about increased vulnerability of the very young or old is very convincing. It has also been reported that male patients are more prone to infection compared to female patients (Mertens et al., 1994; Velasco et al., 1998; Emori et al., 2000; Geubbles et al., 2000; Topal et al., 2000) although the weight of evidence is less strong than that, for age. Individuals with more severe illness and multiple diagnoses are more likely to develop an infection. The more frequently found illness was diabetes mellitus (HIC, 1986; CDC, 1991; Haley, 1991; Pittet & Ducel, 1994; Mertens et al., 1994; Haley, 1995; Velasco et Chapter 2 Literature Review

28

al., 1996; Salemi et al., 1997; Velasco et al., 1998; Pryor & Messmer, 1998; Taylor et al., 1998; de-Boer et al., 1999; Emori et al., 2000; Gaynes, 2000; Vilar-Compte et al., 2000; Soichiro et al., 2000; Denis et al., 2000; Sohn et al., 2000; Geubbles et al., 2000; Topal et al., 2000; Alvarez et al., 2000). Obesity and malnutrition also are considered to have an impact on wound infection (Nichols, 1992; Gerard et al., 1995; David et al., 1995; Moro et al., 1996; Taylor et al., 1998; Bertin et al., 1998; Velasco et al., 1998; Vilar-Compte et al., 2000). Joffe et al., (2000), found that obesity was associated with an increased risk of surgical site infection. Operation related factors An important operation related factor is the surgical wound class. Rates of surgical wound infection increases as the classification of operations progress from clean to dirty wounds (HIC, 1986; CDC, 1991; Nichols, 1993; Pittet & Ducel, 1994; Gerard et al., 1995; Leaper et al., 1996; Arto et al., 1997; Lizan-Gracia et al., 1997; Orrett et al., 1998; Velasco et al., 1998; Taylor et al., 1998; Carl et al., 1999; de-Boer et al., 1999; Kirkland et al., 1999; Emori et al., 2000; Gaynes, 2000; Sohn et al., 2000; Geubbles et al., 2000). Multiple incisions through the same site also enhances the chance for wound infection (Nichols, 1992; Moro et al., 1996; de-Boer et al., 1999; Emori et al., 2000). Emergency surgery was found to be another cause of increased surgical wound infection (HIC, 1986; Gil-Egea et al., 1987; Taylor et al., 1998; Emori et al., 2000; Sohn et al., 2000). There is considerable evidence that the duration of surgery plays an important role in wound infection. The longer the duration the greater the chances for wound infection (HIC, 1986; CDC, 1991; Nichols, 1992; Ferraz et al., 1992; Pittet & Ducel, 1994; Moro et al., 1996; Velasco et al., 1996; Lizan-Gracia et al., 1997; Arto et al., 1997; Orrett et al., 1998; Velasco et al., 1998; Taylor et al., 1998; Pryor, 1998; NNIS, 1999; Kirkland et al., 1999; Emori et al., 2000; Gaynes, 2000; Geubbles et al., 2000; Ganguly et al., 2000). Open surgical drains lasting more than three days have also been identified as increasing the chances of wound infection (HIC, 1986; Nichols, 1992; Moro et al., 1996; Kirkland et al., 1999; Vilar-Compte et al., 2000; Sohn et al., 2000; Ganguly et al., 2000).

Chapter 2 Literature Review

29

Organisational factors The length of pre-operative stay in hospital has a statistically significant association with wound infection. The longer the pre-operative stay, the greater the chances of wound infection (HIC, 1986; Li Liu-yi & Wang, 1990; Mishriki et al., 1990; Mertens et al., 1994; de-Boer et al., 1999; Velasco et al., 1996; Leaper et al., 1996; Lizan-Gracia et al., 1997; Orrett et al., 1998; Taylor et al., 1998; Kirkland et al., 1999; Gaynes, 2000; Ganguly et al., 2000; Topal et al., 2000). One practice considered to contribute to intra-operative microbial contamination is preoperative hair removal (Cruse & Foord, 1996; Mishriki et al., 1990; Nichols, 1992; Small, 1996; Ayliffe, 1996). Shaving of the operation site more than two hours before the procedure relates to a higher infection rate (Leaper et al., 1996). Pre-operative hair removal with a razor was found to increase the risk for surgical wound infection compared with using a clipper. Sellick, et al., (1991) found that there was a significant decrease in surgical wound infection from 1.2% to 0.2% when clippers were used instead of razors. Pre-operative hair removal with a razor is proved to be an important factor in increasing the chances for post-operative wound infection (Mishriki et al., 1990; Nichols, 1992; Hagen et al., 1995; Moro et al., 1996; Siddiqui et al., 1997; Taylor et al., 1998). Health personnel related factors A strong association between the surgeon and the development of wound infection was demonstrated by various researchers (HIC, 1986; Mishriki et al., 1990; Medina-Cuardos et al., 1996; Siddiqui et al., 1997; Holzheimer et al., 1997; Roy & Perl, 1997). Other factors identified were poor hygiene among health personnel, and operation theatre personnel moving in and out of the theatre after they have changed and washed (Siddiqui et al., 1997; Orrett et al., 1998; Flavia & Patricia, 1998; Khuri-Bulos et al., 1999). The operating room environment, ventilation systems, cleansing and sterilisation, and operating room personnel are other related causes for surgical wound infection (HIC, 1986; Pittet & Ducel, 1994). Flavia et al., (1998) in their study observed a rise in surgical site infection as the number of people increased in the operation room. Operating room traffic was identified as a risk factor in surgical site infection (Clarvit et al., 2000).

Chapter 2 Literature Review

30

The indiscriminate use of antibiotics has changed the distribution of pathogens isolated from surgical wound infections. An increasing proportion of surgical wound infections are caused by MRSA (Malcolm, 1997; Mangram et al., 1999). A retrospective study done in one of the cancer hospitals in India also revealed a sudden climb of MRSA from 17% in 1988 to 30% in 1996 as a cause of surgical wound infection, and from 8% to 30% in blood stream infections (Kelkar, 1997). Inappropriate use of antibiotic administration has been identified as a contributory factor to surgical wound infection (HIC, 1986; LizanGarcia et al., 1997; Orrett et al., 1998; Taylor et al., 1998; Ganguly et al., 2000; Gottrup, 2000; Ojha & Deodhar, 2001). However proper use of prophylactic antibiotics administration has proven to be effective in reducing surgical wound infection (Dellinger, et al., 1994; David, 1994; Thomas & Rotstein, 1994; Martin, 1994; Holzheimer et al., 1997; Morikane et al., 2000). Velasco et al., (1998) used a multivariate stepwise logistic regression model to identify independent predictive risk factors causing surgical wound infection. The factors identified were contaminated and infected operations, surgical duration greater than 280 minutes, male sex, prior radiotherapy, American Society of Anesthesiology class III to V and antimicrobial prophylaxis not according to protocol (Velasco et al., 1998). The univariate analysis done by Bertin et al., (1998) showed that there was a significant association between surgical site infection and obesity (78% of cases versus 40% of controls, p=0.02) and older age (mean age 66 years and for cases versus 56 years for controls, p= 0.005). The influence of multiple risk factors can be assessed by performing multiple linear regression analysis (Richards, 2000). The use of stepwise logistic regression analysis to find out the factors contributing to surgical wound infection identified that the condition of the patient prior to surgery and the surgeon performing the surgery were independent risk factors (Medina et al., 1997). Logistic regression model for multivariate analysis of the relative importance of risk factors for surgical wound infection showed that age, prolonged pre-operative stay, wound contamination class, anaesthesia score and duration of surgery were independent risk factors for surgical wound infection (Geubbles et al., 2000)

Chapter 2 Literature Review

31

Multiple analysis of ASEPSIS scores indicated that operation type, ward, degree of contamination, age, body mass index, and pre-operative stay in hospital were significant risk factors for surgical wound infection (Wilson et al., 1990). By performing multiple logistic regression analysis, Arto et al., (1997) identified that alcohol abuse (p 70

60-69

50-59

40-49

30-39

20-29

15

24

2.4%

14

1.4%

Table 23. Distribution of group 1 and group 2 sample according to pre-operative shaving time Variable Pre-

Group 1

Group 2

Number

Percentage

Mean

Range

Number

Percentage

Mean

Range

324

32.4%

6.1

0.5 to

277

27.7%

7.2

0.5 to

hour

24

hours

25

operative shaving time < 1 hour

hours

hours

1-5 hours

284

28.4%

341

34.1%

6-10 hours

94

9.4%

16

1.6%

> 10 hours

298

29.8%

366

36.6%

The mean pre-operative shaving time, ie the time between shaving and operation was greater in group 2 (7.2 hours) when compared to group 1 (6.1 hours) These results are presented in Table 23. In group 2 majority of the patients were shaved more than ten hours prior to surgery. This may be due to the fact there was a shortage of nurses during the post implementation period of data collection. Students were withdrawn from the wards as they had their university examination followed by holidays. Therefore staff nurses had to

Chapter 6 Comparison of pre and post intervention results

125

prepare the patients when the ward was comparatively quiet. The mornings were very busy and it was not possible to prepare all the patients on the day of surgery. Category B This section provides the difference in hand washing and wound dressing score obtained by nurses after the intervention. The second part provides qualitative analysis of field notes and observations made after the intervention and second phase of data collection. Nurse variables The same nurse population participated in the pre and post intervention phases of the project (chapter 4). Figure 17. Hand washing score of the pre and post intervention group. 25

Frequency Distribution

20

15

10

5

Post 0 31

67

20

87

Total

102

58

160

Knowledge score

Nurses who attended the in service education program had a high knowledge score compared to nurses who did not attend the program. Most of the nurses (65.7%) of nurses who attended the program had a score above 31 and only 34% of nurses who did not attend the program had a score above 31 (total score was 40). The findings were statistically significant at 0.001 level. This result clearly shows that education program helped to increase the knowledge.

Appendices

245

APPENDIX 16 Characteristics of group 1 and group 2 patients’ sample Comparison group 1 and group sample by age, pre-operative stay, pre-operative preparation and duration of surgery Variable

Group 1

Group 2

t

p

Age 39.196 ± 17.405 36.73 ± 16.025 1.45 0.130 Pre-operative stay 3.114 ± 4.126 2.86 ± 3.709 4.98 0.148 Pre-operative preparation 6.646 ± 7.3 7.6 ± 7.56 2.873 0.004 Duration of surgery 1.533 ± 0.864 1.636 ± 0.864 1.73 0.06 Except pre-operative preparation all other variables were not significant at 1% level. Comparison of group 1 and group 2 samples by gender Variable

Group 1

Group 2

χ2

df

p

Male 504 559 Female 496 441 Total 1000 1000 4.07 1 0.045 There was no significant gender difference in group 1 and group 2 at 1% level. Comparison of group 1 and group 2 samples by co-morbidity Variable

Group 1

Group 2

χ2

df

p

Co-Morbidity 75 52 No co-morbidity 925 948 Total 1000 1000 4.448 1 0.035 There was no significant difference in co-morbidity in group 1 and group at 1% level.

Appendices

246

Comparison of group 1 and group 2 samples by type of surgery Variable

Group 1

Group 2

χ2

df

p

Emergency 172 137 Elective 828 863 Total 1000 1000 4.689 1 0.03 There was no significant difference in type of surgery in group 1 and group 2 at 1% level. Comparison of group 1 and group 2 samples by wound class Variable

Group 1

Group 2

χ2

df

p

Wound class 1 558 559 Wound class 2 215 242 Wound class 3 152 157 Wound class 4 75 42 Total 1000 1000 7.98 3 0.050 There was no significant difference in wound class in group 1 and group 2 at 1% level. Operations performed by each surgeon in group 1 and group 2 Variable; Surgeon

Group 1

1 123 2 230 3 57 4 131 5 82 6 92 7 165 8 62 9 14 10 8 11 34 12 2 Total 1000 There is no difference in number

Group 2

χ2

df

p

140 223 94 90 76 83 167 84 7 9 23 4 1000 0.612 1 0.424 of operations performed by each surgeon in both the

groups at 1% level of significance

Appendices

247

Comparison of group 1 and group 2 samples by time of surgery Variable

Group 1

Group 2

χ2

df

p

Morning 746 727 Afternoon 254 273 Total 1000 1000 0.93 1 0.335 (Not Significant) There is no difference in number in time of surgery in both the groups at 1% level of significance

Appendices

248

APPENDIX 17 Association between surgical wound infection and variables (ANOVA) Source of

Sum of

variation

Square

Group 1 &

df

Mean

Variatio

sum of

n of

square

ratio

Table df

Inference

4015.15

1

4015.778

122.835

3.846

Significant

Gender

1373.461

1

1373.461

42.012

3.846

Significant

Age

7263.127

6

1210.521

37.028

2.103

Significant

Co-morbidity

25662.98

1

25662.98

784.99

3.846

Significant

Wound class

59957.91

3

18985.97

580.748

.609

Significant

Duration of

220.273

5

44.055

1.348

2.219

Not significant

6.378

1

6.378

.195

3.846

Not significant

72.947

1

72.947

2.231

3.846

Not significant

7815.395

3

2605.132

79.686

2.609

significant

7753.731

4

1938.433

59.293

2.377

Significant

63586.51

1945

32.69

174728.5

1971

Group 2

surgery Emergency/ elective surgery Time of surgery (am/pm) Pre-operative stay Pre-operative preparation time Within group (error) variation Total

The above ANOVA Table shows that except duration of surgery, type of surgery and time surgery all other variables made a significant contribution to surgical wound infection.

Appendices

249

APPENDIX 18 Logistic Regression analysis (2000 sample) to predict significant risk factors Dependent Variable Encoding (Wound infection) Original Value Internal Value .00 0 1.00 1 Categorical Variables Codings Frequency

Paramete (1)

Surgery time am pm Elective Emergency

1743 527 1691 309

1.000 .000 1.000 .000

Gender Male Female

937 1063

1.000 .000

Block 1: Method=Backward Stepwise (Wald) Omnibus Tests of Model Coefficients Chi-square df Sig. Step 1 Step 361.516 9 .000 Blcok 361.516 9 .000 Model 361.516 9 .000 Step 5 Step -1.181 1 .277 Blcok 359.314 5 .000 Model 359.314 5 .000 A negative Chi-square value indicates that the Chi-squares value has decreased from the previous step. Model Summary Step -2Log likelihood 1 763.304 5 765.505

Appendices

Cox & Snell R Square .165 .164

Negelkerke R Square .384 .382

250

Classification Table Predicted

Step 1

INFL

Step 1

Overall Percentage INFL

.00 1.00

INFL .00 1746 172

1.00 18 64

.00 1.00

1744 172

20 64

Overall Percentage

Percentage correct 99.35% 67.90 96.80% 99% 67.90 96.70%

Variables in the Equation Step 1

Step 5

Gender(1) Age Co-morbidity Wound class Duration of surgery Start time(1) Emergency/elective (1) Pre-operative preparation Pre-operative stay Constant Age Co-morbidity Wound class Pre-operative preparation Pre-operative stay Constant

B .064 .023 1.233 .858 .114 .259 .063 .083 .046 -7.963 .024 1.201 .835 .079 .051 -7.640

S.E .217 .006 .273 .095. .115 .251 .306 .017. .021 .697 .006 .273. .092 .016 .020 .634

Wald .086 14.042 20.366 80.669 .982 1.067 .043 24.402 4.692 130.671 15.801 19.629 81.875 23.846 6.402 145.255

df 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Sig .769 .000 .000 .000 .322 .302. .836. .000 .030 .000 .000 .000 .000 .000 .011 .000

Exp(B) .938 1.023 3.431 2.358 1.121 1.295 .939 .920 1.047 .000 1.024 3.354 2.036 .924 1.052 .000

Variable(s) entered on step 1: Age, Gender, Co-morbidity, Wound class, Emergency/elective, Start time, Duration of surgery, Pre-operative preparation, Preoperative stay. Variables not in the Equation Step 5

Variables

Overall Statistics

Appendices

Gender (1) Emergency /elective (1) Start time (1) Duration of surgery

Score

df

Sig.

.030 .060 1.161 1.047 2.222

1 1 1 1 4

.862 .806 .281 .306 .695

251