|Year : 2017 | Volume
| Issue : 3 | Page : 369-375
Antimicrobial prescribing patterns of surgical speciality in a tertiary care hospital in India: Role of persuasive intervention for changing antibiotic prescription behaviour
Chand Wattal1, Shilpi Khanna1, Neeraj Goel1, Jaswinder Kaur Oberoi1, BK Rao2
1 Department of Clinical Microbiology and Immunology, Sir Ganga Ram Hospital, New Delhi, India
2 Department of Critical Care, Sir Ganga Ram Hospital, New Delhi, India
|Date of Web Publication||12-Oct-2017|
Department of Clinical Microbiology and Immunology, Sir Ganga Ram Hospital, New Delhi - 110 060
Source of Support: None, Conflict of Interest: None
Background: Inappropriate use of antibiotics globally has been linked to increase in antibiotic resistance. Objectives: This interventional study assessed the impact of antibiotic prescription feedback and focus group discussions (FGD) on hospital-based prescribers before and after the FGD. Study Design: The present study was performed at a tertiary care centre in New Delhi, wherein 45 units from surgical specialities were included for FGD. Thirty-five units were assessed for the antibiotic usage during 12 months pre-intervention and 3 and 6 months post-intervention period. The outcome measured was a change in antibiotic prescription rates reflected as daily defined doses per 100 bed days as defined by the World Health Organisation. Results: Reduction in the level of antibiotic consumption was observed in 15 of 35 units (42.85%) during the 3 months post-intervention period, which was significant (P < 0.05) in 3/35 (8.57%) surgical units. A significant reduction (P < 0.05) was observed for the units of endoscopic gynaecology, super-speciality and transplant surgery units B and C, and orthopaedic unit C during the 6 months period. Decreasing trend (P < 0.05) was observed in 2/35 (5.71%) units during the entire period. Overall reduction of antibiotic consumption (1.88%) was observed, with an increase in the use of low-end antibiotics and a decrease in the use of high-end antibiotics. Conclusion: The present study clearly demonstrates a weak impact of FGD in changing antibiotic prescribing behaviour. Further analysis of the sustainability of FGD and its long-term impact on antimicrobial resistance needs to be evaluated. The effect of continuous educational sessions and multifaceted interventions cannot be ignored.
Keywords: Antimicrobial stewardship, persuasive intervention, surgical specialities
|How to cite this article:|
Wattal C, Khanna S, Goel N, Oberoi JK, Rao B K. Antimicrobial prescribing patterns of surgical speciality in a tertiary care hospital in India: Role of persuasive intervention for changing antibiotic prescription behaviour. Indian J Med Microbiol 2017;35:369-75
|How to cite this URL:|
Wattal C, Khanna S, Goel N, Oberoi JK, Rao B K. Antimicrobial prescribing patterns of surgical speciality in a tertiary care hospital in India: Role of persuasive intervention for changing antibiotic prescription behaviour. Indian J Med Microbiol [serial online] 2017 [cited 2018 Nov 17];35:369-75. Available from: http://www.ijmm.org/text.asp?2017/35/3/369/216628
| ~ Introduction|| |
Increased and inappropriate antibiotic use among outpatients and in hospitals has been noted worldwide, and the association between antibiotic use and the emergence of antibiotic resistance is well-established.,,, Antimicrobial stewardship (AMS) programmes aim to optimise the quality of antibiotic prescribing while minimising levels of antimicrobial resistance. Qualitative and quantitative assessment of antibiotic use is the mainstay of antibiotic stewardship programmes., Quantitative measurement of antibiotic use is usually carried out by the ATC/DDD system recommended by the World Health Organisation (WHO).
In a previous study done at a tertiary care centre, New Delhi, passive feedback intervention was performed for 33 units (surgical and medical units randomised to intervention and control groups) in an attempt to reduce antibiotic consumption. At the end of the study, it was observed that high and low prescribers continued to prescribe high and low, respectively. Therefore, it was concluded that more active measures as training workshops, frontend approach, or focus group discussions (FGD) may need to be implemented to change behavioural practices of prescribers towards antibiotics.
In continuation to the above study, persuasive intervention in the form of monthly prescription auditing and feedback on antimicrobial prescribing, dissemination of educational resources and reminders were continued, but, in addition, FGD were performed. The aim of this study was to evaluate the impact of FGD as an adjunct intervention to the feedback strategy on antibiotic prescribing rates measured as daily defined doses/100 bed days (DDD/100BD) on hospital-based prescribers before and after the FGD.
| ~ Materials and Methods|| |
The study was conducted at a tertiary care centre with 675 beds, in New Delhi, by the Department of Clinical Microbiology and Immunology from June 2013 to August 2015 in association with the Hospital Infection Control Committee. The doctors of 45 units of the hospital were included from surgical (n = 45) (cardiac surgery [n = 1], obstetrics and gynaecology [n = 4], ENT [n = 4], urology [n = 3], general surgery [n = 4], plastic and cosmetic surgery [n = 2], superspeciality surgery [n = 4], neurosurgery [n = 6], orthopaedics [n = 6], ophthalmology [n = 4], thoracic surgery [n = 2], vascular and endovascular surgery [n = 1], bariatric surgery [n = 1], gynae endoscopy and arthroscopy [n = 2] joint replacement surgery [n = 1]) specialties. Medical units were excluded, to be included in the later phase of the study. Furthermore, units, who generally did not prescribe antibiotics (e.g., psychiatrists, dermatologists, doctors practicing alternative medicine, and laboratory-based doctors) or for whom the methodology used does not provide guidelines for measuring antibiotic consumption as paediatric surgery and neonatology units were excluded from the study. Antimicrobial prescribing rates from free wards, intensive care unit, outpatients, day-care wards and dialysis were also excluded from the study. Prescription of a member of the unit was considered a group decision of that unit.
Antibiotic consumption data were retrieved electronically by speedminer software (Petaling Jaya, Malaysia), part of the hospital information system (Intersystem Cambridge, MA, USA) and was expressed in DDD/100BD as defined by the WHO. Interventional feedback in the form of a feedback sheet containing relevant information about their respective units' prescription pattern along with information on their unit's percentile score and the median antibiotic prescribing across all units in the hospital on a monthly basis was distributed to all the doctors of the involved units, electronically and as a hard copy [Figure 1] and [Table 1]. Persuasive intervention in the form of FGD was performed for each unit focussing on their prescription behaviour and changes that could be done to reduce antibiotic consumption in the respective units. Method, the significance of measuring antibiotic consumption, the possible reasons for high antibiotic prescription of a particular unit and adherence to the antibiotic policy were discussed. At the end, consensus recommendations were made with the aim to reduce their antibiotic prescription rate. Every month FGD was conducted with one speciality so as to cover all specialities included in the study on turnkey basis. The month of the FGD with a speciality (consisting of few units with similar demographics) was taken as the month of persuasive intervention for that unit. The entire proceedings of the meeting were compiled and shared with all members of the unit for its exactness before circulating in that speciality. FGD was considered successful even if a single doctor of the unit was present for the meeting. Retrospective antibiotic consumption data from July 2012 to June 2013, for all units, was taken as the baseline antimicrobial prescribing behaviour pattern for the respective units undergoing FGD. Total systemic (oral or parenteral) antimicrobial use in each unit, measured in DDD/100 bed days was taken as the sole outcome measure. To assess the impact and sustainability of FGD, retrospective or preintervention antibiotic consumption data were compared with the prospective or post-intervention data of the respective units 3 and 6 months post-FGD. Institutional consent from the hospital administration and ethical approval (reference no. EC/11/13/608) was obtained before the study.
|Figure 1: Graphical comparison of antibiotic prescription of a unit with their peers and all units of the hospital as depicted in the feedback sheets|
Click here to view
|Table 1: Regression analysis of interrupted time series on changes in antibiotic consumption|
Click here to view
Median values of antibiotic consumption by units were calculated for 1 year pre-intervention and 3 and 6 months post-intervention period. The outcome was assessed using segmented regression analysis of interrupted time series data which is the strongest quasi-experimental design to assess the longitudinal effects of the intervention. This method estimated changes in the level (a jump or drop in the outcome after the intervention, constituting an abrupt intervention effect) and trend (month-to-month change in the antibiotic consumption before as well as after the intervention) for the outcome measure while taking into account the baseline trends of previous year for that particular unit. The value of P < 0.05 was considered statistically significant.
| ~ Results|| |
Of the 45 units included in the study, intervention data were analysed for 35 (77.77%) surgical units of the hospital. Of the 10 (22.22%) units for which intervention data were not analysed, FGD could not be conducted for 1/10 (10%) units, in-patient bed days were insufficient in 5/10 (50%) units to derive a meaningful result and FGD was considered unsuccessful for 4/10 (40%) units as no one turned up for the meeting [Figure 2].
|Figure 2: Flowchart showing inclusion and exclusion criteria of the involved units|
Click here to view
Overall antibiotic consumption of the surgical units
The median DDD/100 BD for all surgical units of the hospital from July 2012 to August 2015 was 187.10 (range 185.88–190.68). Overall antibiotic prescription reduced from 190.68 DDD/100BDs to 185.88 DDD/100 BDs (−2.5%) and 187.14 DDD/100 BDs (−1.88%) in 3 and 6 months period post-intervention, respectively. Increase in the use of penicillin, 2nd generation cephalosporins and clindamycin was associated with a decrease in the use of 3rd and 4th generation cephalosporins, β-lactam inhibitors, quinolones, macrolides, aminoglycosides, carbapenems, glycopeptides, linezolid, colistin and tigecycline [Figure 3]. The unit of endoscopic gynaecology prescribed least antibiotics followed by the units of super speciality and transplant surgery and cardiac surgery.
|Figure 3: Antibiotic consumption (DDD/100BD) for surgical units during pre-and post-intervention periods|
Click here to view
Impact of intervention
To assess the impact of the intervention on antibiotic consumption in the included surgical units during 3 and 6 months post-intervention period, intuitive graphical presentation of results and visual inspection of the series over time was performed. In an effort to standardise for the seasonal variations in antibiotic prescribing, graph was plotted by dividing figures of a month by those of same month next year [Figure 4] and [Table 1].
Post-intervention at 3 months
Wherein a decrease in antibiotic consumption occurred in 15 of 35 units (42.85%), a statistically significant reduction (P < 0.05) was observed in 3/35 (8.57%) units. The units of endoscopic gynaecology, cardiac surgery and super speciality and transplant surgery B showed a 66.47%, 46.06% and 26.36% reduction in antibiotic consumption from the pre-intervention levels, respectively [Figure 5].
|Figure 5: Graphical presentation of quantitative results and visual inspection of series at 3 and 6 months post-intervention|
Click here to view
Post-intervention at 6 months
Post 6 months following intervention 11/35 (31.42%) units demonstrated a reduction in antimicrobial consumption. Ten of thirty-five (28.57%) units had earlier shown a reduction during the 3 months post-intervention period. Orthopaedics unit F was an additional unit in which a decrease was observed during this period. A significant reduction (P < 0.05) was observed for the units of endoscopic gynaecology, super-speciality and transplant surgery units B and C, and orthopaedic unit C with 71.06%, 38.42%, 61.85% and 26.33% decreased usage, respectively. Although 24.09% reduction in antibiotic consumption occurred in the unit of cardiac surgery, it could not attain a significant level (P < 0.05) at 6 months. However, antibiotic prescription increased in the 5/15 (33.33%) units, which had previously shown non-significant decrease in antibiotic consumption [Figure 5].
Sustainability of impact
Sustained impact over time defined as a persistent reduction during the 3 and 6 months period post-intervention with significant levels (P < 0.05) at the end of 6 months persisted in 4/35 units (11.42%) (Endoscopic gynaecology, super-speciality surgery and transplant units B and C and orthopaedics C) indicating poor sustainability of the impact.
Effect on trend
A reducing change in trend, defined by a decrease in the slope of the segment after the intervention as compared with the segment preceding the intervention was observed in 13/35 (37.14%) units. Of these, 7/13 (20%) units had demonstrated a reduction in levels post-intervention either during the 3 or 6 months period. Significant decrease, however (P < 0.05) was observed in 2/35 (5.71%) units (Superspeciality surgery and transplant unit B and Orthopaedics unit E) during the entire post-intervention period. Although slight month to month change was observed in 6/35 (17.14%) units, it would be an overestimation as they already had a downward trend over time during pre-intervention [Table 1].
Reversal of effects/no change
The intervention did not impact 19/35 units (54.28%) as is evident by an increase/no change in the antibiotic consumption during the entire 3 and 6 months post-intervention period.
| ~ Discussion|| |
An effort was undertaken using a combination of passive and persuasive interventions among surgical units of the hospital to see if this could result in the reduction of the antibiotic prescription.
This study highlights two important facts. First, the meagre reduction (−1.88%) in antibiotic consumption post intervention and second, poor sustainability of the impact shown by initial (post 3 months of intervention) reduction of antibiotic prescription seen in 42.8% of the units which further reduced to 31.42% at 6 months of intervention. Although previous intervention studies have demonstrated successful strategies for altering prescribers' behaviour, most have focussed on discouraging use of specific drugs rather than reducing overall antibiotic prescriptions., In contrast, literature from the west has demonstrated that the dissemination of educational materials in the context of focused stewardship activities can be successful. In a Cochrane review published in 2013, dissemination of educational materials through printed forms or meetings was associated with reduced antibiotic use in 5 of 6 studies; the reduction ranging from 10.6% to 42.5%.
The study demonstrates the varied results obtained whereby few units demonstrated a significant decrease in the level of antibiotic consumption while the rest either remained unchanged or demonstrated an increase even after adding an intervention like FGD. Furthermore among units which showed a decrease in DDD/100 BD following 3 months post-intervention, but did not attain a significant level in the majority. The study throws up the challenge at the sustainability of the intervention as the effects did not persist for > 3 months irrespective of the surgical speciality. For example, the unit of cardiac surgery, handling mostly clean surgeries, did have a decrease in the antibiotic consumption by the end of 3 months post-intervention, but the DDD/100 BD soon increased following an isolated case of microbiologically unproven mediastinitis. This caused a decrease in confidence among the treating surgeons due to which they reverted to their pre-intervention practice. It was however agreed on that factors other than a reduction in antibiotic prescriptions lead to this episode and corrective actions were taken. This attribute has also been discussed previously wherein a decrease in confidence has been observed subsequent to the introduction of a new protocol for a prescription., Landgren et al., performed an educational programme that targeted peri-operative prophylaxis as an intervention. This resulted in antibiotic prescription improvement during the intervention period but could not be sustained over the next 12 months.
Although quantitative measurements of antibiotic prescribing have been described, behavioural change is difficult to be quantified. Factors that promote overuse of antibiotics are lack of education, prior experience, patients' expectations and economics. The patients' expectations of being prescribed antibiotics while being admitted, insecurities leading to surgical sepsis faced by surgeons and lack of adequate knowledge and confidence, contribute mostly to misuse and overuse of antibiotics in a tertiary care hospital. Furthermore prescribing is subject to multiple variations due to cultural beliefs, prescribing etiquette resulting from medical hierarchy and professional relationships, uncertainty about the exact diagnosis or treatment, concern about time schedule (time pressure and workload), worries because of experience of patients' misfortune due to infections. Moreover, concern about not having the energy to resist demands, fear of medico-legal problems if the patient deteriorates, fear of being perceived as having done nothing for patients and worries about endangering the doctor–patient relationship are other plausible reasons for non-pharmacological prescribing of antibiotics., As persuasive interventions rely on behaviour change of individual prescribers, an understanding of the barriers and facilitators of behaviour change should be understood first and the interventions should be targeted to the audience in whom behaviour change is desired. In addition, more supervision of antibiotic prescriptions and doses prescribed by junior doctors need be monitored by senior surgeons as mentors.
The inadequate diagnostic tests encourage antibiotic prescriptions for situations where these are not required. Another reason for antibiotic abuse is the cost of antibiotics which is much cheaper than the cost of investigations. Economically weaker sections who cannot afford the cost of treatment and diagnostics both, a choice has to be made between the two.
The major strength of the study is time series designs and analysis to estimate intervention effects. Segmented regression analysis was used to estimate dynamic changes and outcomes following intervention intended to reduce antibiotic use while controlling for secular changes that may have occurred in the absence of intervention. While this study focused on the effect of FGD when the previous passive intervention was allowed to continue in the form of feedback sheets. Therefore, the current results obtained could be as a result of the combination of both the interventions.
Some limitations of the study however exist. First, the study was conducted in a tertiary care hospital with all varieties of referral cases and an active transplant centre, a case mix was observed in most of the surgical units. Deviation of patient categories from clean cases to clean contaminated and contaminated cases can result in the use of antibiotics irrationally. Second, the outcome was measured in terms of reduction of antibiotic consumption. The outcome in terms of clinical cure of patient and infection free days if included would have been a better indicator. It would also be helpful in instilling confidence among the prescribers. Third, segmented regression analysis generally aggregates individual-level data by time point. The unit of analysis was the monthly median prescription data of the entire unit, rather than each's prescription per month. Thus, individual-level covariates were not addressed in the analysis.
However, on follow-up, the antibiotic consumption of all units after the post-intervention period, the DDD/100 bed days did not change even after completing a year of (until December 2016) the time-bound FGD. This further reiterates the poor sustainability of the backend approach as an intervention. Furthermore, short-term impact of the educational methods and discussions were found hard to be sustainable over a longer period. The knowledge base of the prescribers was also a limiting factor in attaining behavioural changes over a period.
| ~ Conclusion|| |
As education and FGDs are the backhand approach in AMS wherein the usage of antibiotics are controlled after their prescription, further studies on the utility of front hand approach wherein the antibiotics are restricted before their prescription by strategies such as pre-authorisation requirements, formulary restrictions and computer decision support systems needs to be tested. Rigorous antibiotic policy implementation followed by restrictive interventions and a multimodality approach is the need of the hour. Multiple FGDs and a prolonged period should be used to measure the impact of the period. It is recommended and imperative that AMS be included in the curriculum during graduate and postgraduate training of the medical students.
We would like to thank Parul Thakkar for assisting in the statistical calculation of the results.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| ~ References|| |
Polk RE, Fox C, Mahoney A, Letcavage J, MacDougall C. Measurement of adult antibacterial drug use in 130 US hospitals: Comparison of defined daily dose and days of therapy. Clin Infect Dis 2007;44:664-70.
Lepper PM, Grusa E, Reichl H, Högel J, Trautmann M. Consumption of imipenem correlates with beta-lactam resistance in Pseudomonas aeruginosa
. Antimicrob Agents Chemother 2002;46:2920-5.
Goel N, Wattal C, Oberoi JK, Raveendran R, Datta S, Prasad KJ, et al.
Trend analysis of antimicrobial consumption and development of resistance in non-fermenters in a tertiary care hospital in Delhi, India. J Antimicrob Chemother 2011;66:1625-30.
Datta S, Wattal C, Goel N, Oberoi JK, Raveendran R, Prasad KJ, et al.
Aten year analysis of multi-drug resistant blood stream infections caused by Escherichia coli
and Klebsiella pneumoniae
in a tertiary care hospital. Indian J Med Res 2012;135:907-12.
] [Full text]
Dellit TH, Owens RC, McGowan JE Jr., Gerding DN, Weinstein RA, Burke JP, et al.
Infectious Diseases Society of America and the Society for Healthcare Epidemiology of America guidelines for developing an institutional program to enhance antimicrobial stewardship. Clin Infect Dis 2007;44:159-77.
MacDougall C, Polk RE. Antimicrobial stewardship programs in health care systems. Clin Microbiol Rev 2005;18:638-56.
WHO Collaborating Centre for Drug Statistics Methodology. ATC Index with DDDs 2007. Oslo: WHO Collaborating Centre; 2006.
Wattal C, Goel N, Khanna S, Byotra SP, Laxminarayan R, Easton A, et al.
Impact of informational feedback to clinicians on antibiotic-prescribing rates in a tertiary care hospital in Delhi. Indian J Med Microbiol 2015;33:255-9.
] [Full text]
Wagner AK, Soumerai SB, Zhang F, Ross-Degnan D. Segmented regression analysis of interrupted time series studies in medication use research. J Clin Pharm Ther 2002;27:299-309.
Schaffner W, Ray WA, Federspiel CF, Miller WO. Improving antibiotic prescribing in office practice. A controlled trial of three educational methods. JAMA 1983;250:1728-32.
Ekedahl A, Andersson SI, Hovelius B, Mölstad S, Liedholm H, Melander A, et al.
Drug prescription attitudes and behaviour of general practitioners. Effects of a problem-oriented educational programme. Eur J Clin Pharmacol 1995;47:381-7.
Davey P, Brown E, Charani E, Fenelon L, Gould IM, Holmes A, et al.
Interventions to improve antibiotic prescribing practices for hospital inpatients. Cochrane Database Syst Rev 2013;(4):CD003543.
Petursson P. GPs' reasons for “non-pharmacological” prescribing of antibiotics. A phenomenological study. Scand J Prim Health Care 2005;23:120-5.
Kumar S, Little P, Britten N. Why do general practitioners prescribe antibiotics for sore throat? Grounded theory interview study. BMJ 2003;326:138.
Landgren FT, Harvey KJ, Mashford ML, Moulds RF, Guthrie B, Hemming M, et al.
Changing antibiotic prescribing by educational marketing. Med J Aust 1988;149:595-9.
Butler CC, Rollnick S, Pill R, Maggs-Rapport F, Stott N. Understanding the culture of prescribing: Qualitative study of general practitioners' and patients' perceptions of antibiotics for sore throats. BMJ 1998;317:637-42.
Sir Ganga Ram Hospital – Clinical Microbiology Newsletter, Vol. 23; 2017. p. 7. Available from: http://www.sgrh.com/publications
. [Last accessed on 2017 Aug 03].
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]