Outpatient antibiotic prescription rate and pattern in the private sector in India: Evidence from medical audit data.

The key objective of this research was to generate new evidence on outpatient antibiotic prescription rate and patterns in the private sector in India. We used 12-month period (May 2013 to April 2014) medical audit dataset from IQVIA (formerly IMS Health). We coded the diagnosis provided in the medical audit data to International Statistical Classification of Diseases and Related Health Problems (ICD-10) and the prescribed antibiotics for the diagnosis to Anatomic Therapeutic Chemical (ATC) classification of World Health Organization (ATC index-2016). We calculated and reported antibiotic prescription rate per 1,000 persons per year, by age groups, antibiotic class and disease conditions. Our main findings are-approximately 519 million antibiotic prescriptions were dispensed in the private sector, which translates into 412 prescriptions per 1,000 persons per year. Majority of the antibiotic prescriptions were dispensed for acute upper respiratory infections (J06) (20.4%); unspecified acute lower respiratory infection (J22) (12.8%); disorders of urinary system (N39) (6.0%); cough (R05) (4.7%); and acute nasopharyngitis (J00) (4.6%) and highest antibiotic prescription rates were observed in the age group 0-4 years. To conclude our study reports first ever country level estimates of antibiotic prescription by antibiotic classes, age groups, and ICD-10 mapped disease conditions.

Introduction

India is considered to be one of the top users of antibiotics. Our previous research had reported that per capita antibiotic consumption in the retail sector in India has increased by around 22%, from 13.1 DID (defined daily dose (DDD) per 1000 inhabitants per day) in 2008 to 16.0 DID, in a span of five years (2012 to 2016).[1] Evidence from another study suggests that between 2000 and 2015, antibiotic consumption increased from 3.2 to 6.5 billion DDDs (103%) while the antibiotic consumption rate increased from 8.2 to 13.6 DIDs (63%) in India.[2] Literature suggests that high burden of infectious diseases could be one of the reasons for high antibiotic use in India. As per the Million Death Study (MDS) diseases of infectious origin such as pneumonia and diarrhea accounted for around 50% (0·67 million of 1·34 million) of all deaths in children aged less than 5 years in India.[3] However, inappropriate use of antibiotics cannot be ruled out.

Although clinical guidelines on judicious antibiotic use[4, 5] explicitly mentions that antibiotics should not be prescribed for common cold, nonspecific upper respiratory tract infection (URI), acute cough illness, and acute bronchitis, literature on antibiotic prescribing from India indicates high rate of antibiotic prescriptions for respiratory infections in primary care.[6-8]. Recent evidence from United States also suggests that seasonal peaks in antibiotics use during cold and influenza season suggestive of viral upper respiratory tract infections antibiotic has remained unchanged highlighting inappropriate prescribing of antibiotics [9]despite release of several antibiotic prescribing guidelines. However, evidence from UK suggest that standardized consultation rate for ‘any respiratory infection’ declined by 35 per cent and overall antibiotic prescriptions for all acute respiratory infections declined by 45 per cent between 1994 and 2000.[10] Another interesting study from Taiwan reported that children with a physician or a pharmacist as a parent were significantly less likely than other children to receive antibiotic prescriptions suggesting better education does help in reducing the frequency of injudicious antibiotic prescribing.[11]

Evidence from India has also highlighted relatively high antibiotic prescription rate in private health facilities as compared to public health facilities.[7, 12] These studies also highlighted the frequent use of expensive newer classes of antibiotics as compared to the older ones in the private sector.[7, 8, 12, 13] One of the possible reasons for such a trend is the dominance of the private sector in funding and provisioning of health care in India, as per the National Sample Survey (NSS) nearly 75% of all outpatient visits and about 62% of hospitalization episodes occurred in private health delivery system in the year 2014.[14] Furthermore, households largely buy medicines directly from retail pharmacies as prescribed by the general practitioners in the private sector.[15]

The health system elements outlined above clearly demonstrate the role and relevance of primary care physician/general practitioner of the private sector in medicine use and in particular antibiotic use. Previous research involving micro level surveys revealed several facets of inappropriate medicine use in the Indian context.[7, 8, 16–18] However, none of them were truly representative of private sector primary care physicians because of their limited sample size and limited geographical locations. We conducted this research to generate new evidence on outpatient antibiotic (J01) prescription rate and patterns in the private sector in India. We also performed additional analysis to report age-specific and disease-specific differences by different antibiotic classes.

Material and methods

Data source and setting

We examined prescription rates and patterns of antibiotics (J01) of primary care physicians working in the private sector in India with the help of IQVIA medical audit data (formerly IMS Health) for a 12 month period (May 2013 to April 2014).[19] IQVIA is a for-profit organisation that collects and provides data and information on pharmaceutical market intelligence in over 100 countries around the world. The medical audit data tracks prescriptions by private practitioners practicing allopathic system of medicine. This data is collected from a panel comprising of 4600 doctors selected through a multi-stage stratified random sampling, which include general practitioners (MBBS, Bachelor of Medicine, Bachelor of Surgery), non-MBBS general practitioners, and other medical specialties (such as dentists, pediatrics, gynecology, dermatology, and others) from 23 metropolitan areas (population more than 1 million), 128 class 1 towns (population more than 100,000) and 1A towns (population less than 100,000) of India. The data is then extrapolated to reflect the prescription pattern of doctors having private practices in towns with population more than a hundred thousand across the country.

This database provides information on patient characteristics such as age, gender, symptoms, diagnosis and the medicines prescribed. The data organizes medicines according to anatomical therapeutic classification (ATC) of the European Pharmaceutical Market Research Association (EphMRA) but not according to the World Health Organisation’s ATC classification. Also, the diagnosis reported on prescriptions are not coded to the International Statistical Classification of Diseases and Related Health Problems (ICD-10). Furthermore, the data does not capture the public sector prescriptions and therefore our analysis only reflects outpatient antibiotic prescription patterns in the private sector in the country.

Finally, the data was made available to us by IQVIA in the form of aggregates processed and extrapolated to reflect the prescription practices in the country instead of being made available in raw form as individual-level data. The data we used had no identifiers for the patients. We therefore did not require ethical approval for our study.

Outcome measure

Our primary outcome measure was antibiotic prescription rate per 1,000 persons per year. We also estimated and reported age-specific and disease-specific antibiotic prescriptions by antibiotic classes.

Statistical analysis

We coded the diagnosis provided on the prescription in the IQVIA medical audit data to disease classifications based on the International Statistical Classification of Diseases and Related Health Problems, 10th Revision (ICD-10 classification; version: 2016)[20] and the antibiotics prescribed for the related diagnosis to the 3rd level of Anatomic Therapeutic Chemical (ATC) classification as per the methodology proposed by World Health Organization’s Collaborating Centre (WHOCC) of Drug Statistics Methodology’s (ATC index-2016).[21] The diagnosis was categorized into ICD 10 codes through a search on the online index using specific key words in the diagnosis provided in the medical audit data, which was taken directly from the prescriptions. The idea was to code the ICD 10 codes up to the narrowest (most detailed) level possible depending on the extent of details on the diagnosis provided in the medical audit data.

The utilization of antibiotics (ATC code: J01) was measured in terms of the annual prescription rate, i.e. number of antibiotic prescriptions divided by 1000 person years. The population estimates were obtained from the report of the technical group on population projections constituted by the National Commission on Population.[22] Age-groups were determined by the classification already provided in the medical audit data. The medicines prescribed were classified into the following antibiotic subgroups (ATC codes): tetracyclines (J01A), amphenicols (J01B), penicillins (J01C), other beta-lactams, cephalosporins (J01D), sulfonamides & trimethoprim (J01E), macrolides, lincosamides and streptogramins (J01F), aminoglycosides (J01G), Quinolones (J01M), combinations of antibacterials (J01R), other antibacterials (J01X). Unclassifiable antibiotics were pooled in the subgroup ‘others’.

We used per 1000 persons as denominator in contrast to individuals because the data was available only for prescription per person and not for an individual for the entire year. Antibiotic prescription rate is a better indicator for antibiotic use[23] as compared to defined daily doses (DDD) per person since antibiotic dose depends on a patient’s age and body weight. We analyzed and reported antibiotic use by age-groups and disease conditions (based on ICD-10 classification) expressed as annual prescription rate per 1,000 persons for each class of antibiotics. We used software STATA 14.0 to perform statistical analysis.

Results

Antibiotic prescribing pattern

We present new evidence on outpatient antibiotic prescription rates and pattern in the private sector in India. Around 519 million antibiotic prescriptions were dispensed in 2014, which translate into 412 prescriptions per 1,000 persons per year. The antibiotic prescription rates were highest for children aged 0–4 years (636 prescriptions per 1,000 persons) and lowest in the age group 10–19 years (280 prescriptions per 1,000 persons) (Fig 1 and S1 Table). It may also be noted that across all age groups, beta-lactam, cephalosporins (J01D) had the highest prescription rates (38.3% of all antibiotic prescriptions) followed by beta-lactam, penicillins (J01C) (22.8%) and quinolones (J01M) (16.3%). It may be further noted that cephalosporins (J01D) were the most commonly prescribed antibiotic across all diagnoses with the exception of disorders of urinary system where quinolones (J01M) were more commonly prescribed.

Fig 1

Outpatient antibiotic prescription rate per 1000 persons per year, by age groups and antibiotic classes, India (2013–2014).

Antibiotic prescribing across clinical diagnoses

Of 519 million antibiotic prescriptions, majority were dispensed for the diseases of the respiratory system (55%), followed by diseases of the genitourinary system (10%), and symptoms, signs and abnormal clinical findings (9%) (Table 1).

Table 1

Distribution of outpatient antibiotic prescriptions in India, by disease conditions, 2013–2014.

ICD chapter number	ICD chapter name	Number of prescriptions	Percentage (%)	Prescription rate per 1000 persons per year
1	Certain infectious and parasitic diseases	16,820,293	3.24	13.35
3	Diseases of the blood and blood-forming organs	1,158,125	0.22	0.92
4	Endocrine, nutritional and metabolic diseases	4,566,575	0.88	3.63
5	Mental and behavioral disorders	967,392	0.19	0.77
6	Diseases of the nervous system	1,491,169	0.29	1.18
7	Diseases of the eye and adnexa	3,685,540	0.71	2.93
8	Diseases of the ear and mastoid process	8,653,317	1.67	6.87
9	Diseases of the circulatory system	8,472,213	1.63	6.73
10	Diseases of the respiratory system	286,059,212	55.09	227.10
11	Diseases of the digestive system	28,532,251	5.49	22.65
12	Diseases of the skin and subcutaneous tissue	20,917,567	4.03	16.61
13	Diseases of the musculoskeletal system	5,973,066	1.15	4.74
14	Diseases of the genitourinary system	51,791,862	9.97	41.12
15	Pregnancy, childbirth and the puerperium	977,532	0.19	0.78
17	Congenital malformations, deformations	114,892	0.02	0.09
18	Symptoms, signs and abnormal clinical and laboratory	44,506,708	8.57	35.33
19	Injury, poisoning and external causes	33,398,837	6.43	26.52
20	External causes of morbidity and mortality	37,318	0.01	0.03
21	Factors influencing health status	1,118,282	0.22	0.89
	Total	519,242,151	100.00	412.23

As per the ICD-10 classification, the following top ten disease conditions contributed approximately 63% of the total antibiotic prescriptions—acute upper respiratory infections (J06) (20.4%; range—12.1% to 25.4%), unspecified acute lower respiratory infection (J22) (12.8%; range—9.0% to 24.7%), disorders of urinary system (N39) (6.0%; range—1.7% to 9.0%), cough (R05) (4.7%; range—2.9% to 6.1%), acute nasopharyngitis (J00) (4.6%; range—2.0% to 6.1%), acute pharyngitis (J02) (3.9%; range– 1.9% to 5.0%), acute bronchitis (J20) (3.4%; range—2.4% to 5.1%), injury, poisoning and others (T14) (2.5%; range– 1.8% to 3.2%), cutaneous abscess and furuncle (L02) (2.3%; range—1.4% to 2.6%), and asthma (J45) (2.2%; range—1.8% to 4.4%) (Fig 2 and S2 Table). (For full information on antibiotic prescription percentages across top ten clinical diagnosis and age groups and across antibiotic classes see supplementary file (S1, S2 and S3 Tables)).

Fig 2

Number of antibiotic prescriptions, by disease conditions and antibiotic classes, India (2013–2014).

Discussion

To the best of our knowledge, our study provides the first ever estimates of outpatient antibiotic prescription rates and patterns in the private sector in towns with population more than a hundred thousand across the country. Our findings illustrate significant variations in antibiotic prescription rates across age groups, by disease conditions (ICD-10 classification) and by antibiotic classes (ATC classification). Earlier Von Bockel et al. and Klein et al. had reported estimates of antibiotic consumption in India through use of pharmaceutical sales data[2, 24]. However, their study did not provide information on antibiotic use by disease conditions and age groups. While other prescription analysis based studies from India reported antibiotic utilization in public and private sector health facilities[7, 8, 12, 13, 16, 25], they had limited sample sizes and geographical locations.

Our estimates suggest high proportion of antibiotic prescription for upper respiratory tract infections (acute upper respiratory infections (J06) (20.4%), cough (R05) (4.7%), acute nasopharyngitis (J00) (4.6%), and acute pharyngitis (J02) (3.9%)). Generally, these infections are viral in origin and are self-limiting in nature. Hence, in the light of evidence based medicine and standard treatment guidelines, it may be argued that a significant proportion of these antibiotic prescriptions might be inappropriate in nature. Previous research on prescription practices also highlighted the problem of inappropriate use of broad-spectrum antibiotics in India. For example, Kotwani et al. had reported that in the private sector, not only were the antibiotic prescription rates higher but the choice of antibiotics for the treatment of uncomplicated respiratory infections too, was inappropriate.[8] Chandy et al reported widespread use of fluoroquinolone, especially by general practitioners[12] in the private sector. Other studies have also reported high cephalosporin use in urban hospitals and pharmacy shops.[12, 26]

Bianco et al, who studies antibiotic prescription to adults with ARTI by Italian GPs, concluded that there was a very high frequency of non evidence-based prescriptions of antibiotics at primary care level. 65.5 percent of times the prescriptions were not being indicated by guideline. [27]Studies have also highlighted the incorrect perception among patients and parents of pediatric patient that antibiotics work for treating viral infections.[28, 29]Napolitano et al reported that only 9.8 percent of the Italian patients surveyed knew the definition of antibiotic resistance and only 21.2 percent knew when it was appropriate to use antibiotics.[30]

Previous research also shows that besides the lack of awareness, inappropriate antibiotic use is linked to supply-side incentives, which lead to over prescription of antibiotics in the private sector.[16, 25, 31] This problem of inappropriate use of antibiotics gets accentuated multifold because of limited access to care and medicines in the public health system[32] which forces patients to seek care in the private sector. Limited access to medicines in the public sector also results in over the counter purchase of antibiotics, which is a major driver of inappropriate use in India. Laxminarayanan et al had reported that non-prescription sales of carbapenems in India are among the highest in the world and contribute to growing carbapenem resistance.[33] Although, over the counter access to antibiotics is a complex problem in India since insufficient access and delays in access to antibiotics causes more deaths than antibiotic resistance.,[34] Numerous studies have reported increasing levels of resistance to last resort antibiotics like carbapenem[33, 35, 36]. Evidence also suggests that inappropriate antibiotic use not only has profound impact on antimicrobial resistance[37] but also on treatment cost because of drug resistant organism.[38]

Our analyses suggest that antibiotic prescription rates (412 prescriptions per 1000 persons in 2014) in India are lower than various European nations. For example, antibiotic prescription rates in Italy (957 prescriptions per 1000 persons), Germany (561 prescriptions per 1000 persons), UK (555 prescriptions per 1000 persons), Denmark (481 prescriptions per 1000 persons)[39] and Greece (1100 antibiotics per 1000 person)[26] are much higher than India. However, antibiotic prescription rates for certain antibiotic classes are on a higher side in India as compared to the developed nations. For example, the percentage of prescriptions with cephalosporins and quinolones (38.2% and 16.3%) in India were significantly higher than USA (14.0% and 12.7%)[40], and Greece (32.9% and 0.5%).[26] Such unusually high prescription rates of beta-lactams-penicillins and cephalosporins in uncomplicated upper respiratory infections in children is in stark contrast to the prescription rates and pattern reported in European countries.[39]

The potential reasons for high prescription rates of broad spectrum antibiotics like cephalosporins and quinolones are not only high burden of infectious diseases, but also lack of diagnostic support services and inadequate training of physicians.[12] Literature also suggests that perceived demand and expectations from the patients, influence from medical representatives and inadequate knowledge influences doctors decisions to prescribe antibiotics.[41] Previous research has highlighted that inappropriate use of medicines is rampant in less regulated health markets[42] and can take several forms: overuse, underuse, misuse, and unnecessary expensive use. Our findings for antibiotic prescription are consistent with antibiotic sales in India. Our previous analysis on pharmaceutical sales data had also suggested that antibiotic consumption (16.0 DID) in India was significantly below the mean antibiotic consumption (21.5 DID) of European countries.[1]

To address the problems related to inappropriate use of antibiotics, government of India has deployed a multipronged strategy. This includes setting up treatment guidelines for antimicrobial use[5] and multi-centric surveillance for tracking antimicrobial resistance[43]. In addition, over-the-counter (OTC) sales of 3rd and 4th generation antibiotics are now regulated through Schedule H1 of Drugs & Cosmetics Rules.[44] Furthermore, to reduce burden of pneumonia and diarrhea and the demand for antibiotics, new vaccines have been introduced in the universal immunization program[45]. However, success of these strategic measures in terms of achieving intended objectives is yet to be demonstrated. Our analysis suggest around 100 million prescriptions were dispensed for acute upper respiratory tract infections alone, antibiotic stewardship programs directed towards diagnosis and treatment of URI could result into significant reduction in antibiotic use.

Our study has certain limitations. The scope of our study was limited to analysis of antibiotic prescription pattern in the private sector in in towns with population more than a hundred thousand across the country, as we did not have access to public sector prescription data. Therefore, the study is not representative of the prescriptions generated in public sector facilities. This may have resulted in an underestimation of antibiotic prescription rates albeit only marginally, since more than 80% of the population seeks care in private sector and approximately 90% of medicine expenditure occurs in private sector. In addition, a small proportion (<1%) of antibiotic prescriptions sometimes got mapped to completely unrelated diagnosis because IQVIA medical audit data is coded in such a way that antibiotics on prescriptions gets mapped to every differential diagnosis (related or completely unrelated) on the prescription. The diagnosis provided in the dataset were not already mapped to ICD 10 classification, therefore the authors had to map the diagnosis to ICD 10 classification based on available information which may have led to certain inaccuracies in allocation of codes as well as the resulting analysis. Since we did not have information on the number of patients accessing general practices in metros and class 1 and 1a towns, we had to rely on the total population to arrive at the prescription rates. This is another limitation of our study.

Conclusions

This research work provides the first ever estimates on antibiotic prescription rate and pattern in the outpatient general practice in the private sector of towns with population more than a hundred thousand across the country. Overall antibiotic prescription rates in India are still much lower than Europe. However, prescription rates for broad-spectrum beta-lactam antibiotics are much higher as compared to European nations, especially in children. Approximately one-fifth antibiotic prescriptions were dispensed for upper respiratory infections, which rarely require an antibiotic therapy. Our findings highlight that primary care physicians in the private sector can play a key role in reducing antibiotic misuse and overuse. Our research findings also provide critical information to target antimicrobial stewardship programs to specific constituencies and stakeholders. This baseline information can also be used as a benchmark for measuring the impact of current and future interventions directed towards reducing inappropriate antibiotic use.

Antibiotic prescription rate per 1000 person per year, by age group, India (2013–2014).

(DOCX)

Click here for additional data file.

Top 10 diagnoses for antibiotic prescriptions, by age group, India (2013–2014).

(DOCX)

Click here for additional data file.

Distribution of outpatient antibiotic prescriptions by disease conditions and antibiotic classes, India (2013–2014).

(DOCX)

Click here for additional data file.

21 Jun 2019

PONE-D-19-14711

Outpatient antibiotic prescription rate and pattern in the private sector in India: Evidence from medical audit data

PLOS ONE

Dear Dr. Farooqui,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

We would appreciate receiving your revised manuscript by July 23, 2019. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Italo Francesco Angelillo, DDS, MPH

Academic Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. In ethics statement in the manuscript and in the online submission form, please provide additional information about the patient records used in your retrospective study. Specifically, please ensure that you have discussed whether all data were fully anonymized before you accessed them and/or whether the IRB or ethics committee waived the requirement for informed consent. If patients provided informed written consent to have data from their medical records used in research, please include this information.

3.  We noticed your Discussion has some minor occurrence of overlapping text with the following previous publication, which needs to be addressed:

Farooqui, Habib Hasan, et al. "Community level antibiotic utilization in India and its comparison vis-à-vis European countries: Evidence from pharmaceutical sales data." PloS one 13.10 (2018): e0204805.

In your revision ensure you cite all your sources (including your own works), and quote or rephrase any duplicated text outside the methods section. Further consideration is dependent on these concerns being addressed.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Dr. Farooqui and colleagues present a descriptive analysis of private-sector antibiotic prescribing in India using IQVIA medical audit data from May 2013-May 2014. The study presents descriptive findings that might be of interest to clinicians and public health practitioners. I felt the analysis did not go very deep into the data, although the discussion nicely highlighted key findings. The manuscript could benefit from a deeper analysis of the data, if possible given the dataset. A few comments for consideration:

• As the diagnoses in your dataset are not in ICD10 codes, how did you categorize them according to ICD10 codes? It would be helpful to include more on this methodology, perhaps consider including a crosswalk in your supplementary materials.

• Why are there no variance estimates? Is this due to the dataset projection methodology? If possible, variance estimates should be included.

• It would make your manuscript stronger if you dug deeper into the descriptive data with a few additional analyses. For example, what diagnoses are responsible for the most antibiotic prescriptions by age group? Is there a statistically significant difference between agents and diagnoses in different age groups.

• Is there any information on region/geographic area in the dataset. Or provider type (more specifically than general practitioner)? That could be interesting to include and might make the analyses more robust.

• What does the ICD code column in Table 1 show?

• In the discussion, it makes sense that you discuss over the counter antibiotic sales as that is a contributor to inappropriate antibiotic use. However, I feel you could trim this section down a little since it is not the focus of your analysis.

• Please review for grammar and punctuation and ensure all abbreviations are defined at their first use.

• I think it would be fine to say IQVIA instead of IMS Health since that is the current company name.

Reviewer #2: Although this approach is meriting, the issues of the paper are not meaningful enough and convincing. I do not agree with the conclusion of the authors: data results could not lead to the conclusion “We observed an inappropriate and high antibiotic prescription rates

for upper respiratory infections in children age less than 5 years”. Indeed, you used a dataset to estimate the rates and the motif of prescription; to prescribe an antibiotic is associated with a lot of arguments and to estimate if appropriate or not you must read the entire medical report, with biological results (you did not include) the symptoms, the X-rays and so on. To use retrospective medical and clinical data to estimate the rate of AB prescription is a tool we have to develop nowadays, but you can’t conclude on pertinence with this kind of methods.

There are different potential methodological reasons for that, the main being that very different clinical presentations could lead to quote a upper respiratory infections; but if the person has comorbidities or other medical condition, maybe an antibiotic could be necessary. You did not adjust your results on the conditions of the patients. The study could be more relevant in presenting the results without this kind of interpretation that can’t be done.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: Outpatient antibiotic prescription rate_PLOSOne_review.doc

Click here for additional data file.

Submitted filename: Plos review v1.pdf

Click here for additional data file.

16 Sep 2019

Outpatient antibiotic prescription rate and pattern in the private sector in India: Evidence from medical audit data

Response to Reviewers

The authors would like to express their gratitude to the reviewers for their insightful comments that have been extremely useful for us in improving the manuscript.

Reviewer #1:

Dr. Farooqui and colleagues present a descriptive analysis of private-sector antibiotic prescribing in India using IQVIA medical audit data from May 2013-May 2014. The study presents descriptive findings that might be of interest to clinicians and public health practitioners. I felt the analysis did not go very deep into the data, although the discussion nicely highlighted key findings. The manuscript could benefit from a deeper analysis of the data, if possible given the dataset.

A few comments for consideration:

As the diagnoses in your dataset are not in ICD10 codes, how did you categorize them according to ICD10 codes? It would be helpful to include more on this methodology, perhaps consider including a crosswalk in your supplementary materials.

Response: The following has been included in the section ‘Material and Methods’ sub-section ‘Statistical Analysis’ Para 1, page 6 of the manuscript to explain the coding process better:

“The diagnosis was categorized into ICD 10 codes through a search on the online index using specific key words in the diagnosis provided in the dataset which was taken directly from the prescriptions. The idea was to code the ICD 10 codes up to the narrowest (most detailed) level possible depending on the extent of details provided in the diagnosis.”

Reviewer: Why are there no variance estimates? Is this due to the dataset projection methodology? If possible, variance estimates should be included.

Response: We worked with data that was already projected and shared with us by IQVIA (formerly IMS Health). We did not carry out the projection ourselves. We are therefore unable to provide variance estimates.

Reviewer: It would make your manuscript stronger if you dug deeper into the descriptive data with a few additional analyses. For example, what diagnoses are responsible for the most antibiotic prescriptions by age group? Is there a statistically significant difference between agents and diagnoses in different age groups.

Response: Thank you for the valuable suggestion. We have now included the table with the top 10 diagnosis for antibiotic prescriptions by age groups in the supplementary material. Please see S2 Table in supplementary materials. This has been referred to in the ‘results’ sections, sub-section ‘Antibiotic prescribing across clinical diagnoses’ para 1, page 10-11.

Reviewer : Is there any information on region/geographic area in the dataset. Or provider type (more specifically than general practitioner)? That could be interesting to include and might make the analyses more robust.

Response: Unfortunately, this information is not available with us. It must be noted that while the data is from private practitioners, it is not only from GPs but also representative of other specializations. We do not however, have information on how the prescription patterns vary across prescriber specializations.

Reviewer : What does the ICD code column in Table 1 show?

Response: The ICD code indicates each of the 21 chapters. This has been re-labeled in table 1 in the manuscript.

Reviewer : In the discussion, it makes sense that you discuss over the counter antibiotic sales as that is a contributor to inappropriate antibiotic use. However, I feel you could trim this section down a little since it is not the focus of your analysis.

Response : As suggested the section on over the counter antibiotic sales has been trimmed.

Reviewer : Please review for grammar and punctuation and ensure all abbreviations are defined at their first use.

Response: The manuscript has been reviewed and revised to address this comment.

Reviewer : I think it would be fine to say IQVIA instead of IMS Health since that is the current company name.

Response: The manuscript has been reviewed and revised to address this comment.

Reviewer #2:

Reviewer: This study proposed to estimate outpatient antibiotic prescription rate and patterns in the private sector in India during one year using one medical audit dataset from IMS Health (now IQVIA). The relevance of this study is the increasing room taken by the big data in healthcare sciences and the switch of paradigm that will be linked to this evolution.

This manuscript exposed their automated method to estimate antibiotic prescription rates and patterns of prescription according to coding practice in a sample of medical reports from the outpatient private sector in India. To conclude the study reported high antibiotic prescription level overall, by antibiotic classes, age groups, and ICD-10 mapped disease conditions.

The topic is highly relevant because antibiotic consumption remains a major public health problem worldwide, promoting the spread of antimicrobial-resistant organisms. Moreover, to measure and survey the antibiotic prescription rate could represent an added value in combination with infection control strategies, for both achieving successful outcomes in patients and impacting the rate of prescriptions.

Although this approach is meriting, the issues of the paper are not meaningful enough and convincing. I do not agree with the conclusion of the authors: data results could not lead to the conclusion “We observed an inappropriate and high antibiotic prescription rates for upper respiratory infections in children age less than 5 years”. Indeed, you used a dataset to estimate the rates and the motif of prescription; to prescribe an antibiotic is associated with a lot of arguments and to estimate if appropriate or not you must read the entire medical report, with biological results (you did not include) the symptoms, the X-rays and so on. To use retrospective medical and clinical data to estimate the rate of AB prescription is a tool we have to develop nowadays, but you can’t conclude on pertinence with this kind of methods.

There are different potential methodological reasons for that, the main being that very different clinical presentations could lead to quote a upper respiratory infections; but if the person has comorbidities or other medical condition, maybe an antibiotic could be necessary. You did not adjust your results on the conditions of the patients. The study could be more relevant in presenting the results without this kind of interpretation that can’t be done.

Response: Thank you for your detailed and useful comments. I agree, the argument around inappropriate use of antibiotics in any clinical diagnosis has to be based not only on symptoms reported by the patient and captured on the prescription but also on sign of fever, laboratory diagnosis and x-rays. However, in India, majority of patients seek care at primary care physicians in the private sector, where they are prescribed empirical antibiotic therapy without any laboratory investigations, microbiologic testing and culture sensitivity. Multiple studies have highlighted these issues and given the epidemiological and microbiological trends and patterns of acute respiratory infections in India, more often, viral aetiology has been identified in community based studies, which do not usually require empirical antibiotic therapy, hence the argument around inappropriate prescription. However, as per your suggestion, the statement has been removed.

Specific comments

Reviewer: The introduction is in my opinion too focused on India, comparisons with outpatient antibiotic prescription in other countries (Western countries but more similar Eastern countries).

Response: Thank you for your valuable comments. Your suggestion has been incorporated in the manuscript in the introduction section. The following has been included in the introduction section (paragraph 2, page 3-4) “Although clinical guidelines on judicious antibiotic…….. does help in reducing the frequency of injudicious antibiotic prescribing”

Method section:

Reviewer: it is well written even if some typos remain (cf. manuscript)

Concerning the representativity of your study population, this sample of GP in the private sector in India.

*how are recruited the GPs? Volontary? At random? Request? Please explain

*how could you be sure they are representative of the entire country?

Response: Details of the sampling strategy have been added in the section ‘Material and Methods’ sub-section ‘Data Source and Setting’ para 1, page 5 of the manuscript. Please see “This data is collected from a panel comprising of 4600 doctors selected through a multi-stage stratified random sampling, which include… »

Further clarification on representativeness has been added in the section ‘Material and Methods’ sub-section ‘Data Source and Setting’ para 1, page 5-6 of the manuscript. Please see « The data is then extrapolated to reflect the prescription pattern of doctors having private practices in towns with population more than a hundred thousand across the country.”

Reviewer: You use the population census as the denominator; however aren’t you afraid of the part of this population going to the public sector of healthcare? And so you used a higher number of persons than the targeted population, giving a misinterpretation of the ab prescription rates? What do you think?

Response: The proportion of patient visiting the public sector is generally very small in India, especially for outpatient care and therefore the private sector provides a good enough representation of the country in general. We have explained this in the discussion section, para 8 page 15-16 of the manuscript « Therefore, the study is not representative of the prescriptions generated in public sector facilities. This may have resulted in an underestimation of antibiotic prescription rates albeit only marginally, since more than 80% of the population seeks care in private sector and approximately 90% of medicine expenditure occurs in private sector »

We have also added the following as a limitation to the discussions section para 8, page 15. “Since we did not have information on the number of the patients accessing general practices in metros and class 1 and 1a towns, we had to rely on the total population to arrive at the prescription rates. This is another limitation of our study.”

Reviewer: The dataset did not have the motif of the visit and you coded yourself in ICD -10 codes? From what kind of database? Who did choose the motif of consultation? The pattern of the visit?

This is a little surprising as you went to specific data to structured but less specific ones???

Explain the initial data before the transcription in ICD-10 codes.

Response: We used the information provided in the IQVIA medical audit dataset on the doctor’s diagnosis on the prescription as the basis for ICD 10 classification. The following has been included in the section ‘Material and Methods’ sub-section ‘Statistical Analysis’ Para 1, page 6-7 of the manuscript to explain the coding process better:

“The diagnosis was categorized into ICD 10 codes through a search on the online index using specific key words in the diagnosis provided in the dataset which was taken directly from the prescriptions. The idea was to code the ICD 10 codes up to the narrowest (most detailed) level possible depending on the extent of details provided in the diagnosis.”

Reviewer: How did you take into account the fact that the person could come twice in the same week for the same event and first not have Ab than as still ill, has finally been prescribed one treatment.

Response: We were unfortunately not able to take this into account due to the nature of our data, which was not individual level. This is a limitation of our study.

Discussion section: cf manuscript

A bit too long even if interesting. Some extra subject paragraphs that could be taken out in my opinion or shortened.

Response: The manuscript has been reviewed and revised to address this comment.

References:

Some typos and lack of recent literature on that topic. Authors should better screened references databases.

Response: The manuscript has been reviewed and revised to address this comment.

Some suggestions to help:

Antibiotic Prescribing in Outpatient Children: A Cohort From a Clinical Data Warehouse. Grammatico-Guillon L, Shea K, Jafarzadeh SR, Camelo I, Maakaroun-Vermesse Z, Figueira M, Adams WG, Pelton S. Clin Pediatr (Phila). 2019 Jun;58(6):681-690.

Durkin MJ, Jafarzadeh SR, Hsueh K, Sallah YH, Munshi KD, Henderson RR, et al. Outpatient Antibiotic Prescription Trends in the United States: A National Cohort Study. Infect Control Hosp Epidemiol. 2018 Feb 27;1–6.

Hersh AL, Jackson MA, Hicks LA, the COMMITTEE ON INFECTIOUS DISEASES. Principles of Judicious Antibiotic Prescribing for Upper Respiratory Tract Infections in Pediatrics. PEDIATRICS. 2013 Dec 1;132(6):1146–54.

Submitted filename: Response to the reviewers.doc

Click here for additional data file.

7 Oct 2019

PONE-D-19-14711R1

Outpatient antibiotic prescription rate and pattern in the private sector in India: Evidence from medical audit data

PLOS ONE

Dear Dr. Farooqui,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

We would appreciate receiving your revised manuscript by October 31, 2019. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Italo Francesco Angelillo, DDS, MPH

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #2: All comments have been addressed

Reviewer #3: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #2: Partly

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #2: No

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #2: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #2: The authors took into account the comments of both reviewers and modified the manuscript accordingly.

This improved the response to the objectives of the study and make the article suitable for publication in my opinion.

Reviewer #3: Many thanks for allowing me to review this revised manuscript.

The authors tried to adequately address the reviewer comments raised in the previous round of review, although they were not able to overcome some shortcomings (e.g. variance estimates, comorbidities and follow-up visits for the same diagnosis were not explored) due to the nature of the study data. However, I believe that inappropriate use of antibiotics is a serious threat to global health, and exploring the pattern of antibiotic prescription among primary care providers represents one of the best tool to promote the appropriateness of use and to control antimicrobial resistance.

The revised manuscript could be approved for publication in PLOS ONE journal, after other minor revision.

I suggest to add in the limitations section that since the dataset did not include the ICD-10 code, the authors coded themselves the ICD-10 starting from the reason of the visit. It may lead to an inaccurate evaluation.

Both the reviewers highlighted lack of recent literature on that topic. All these papers have to be cited and commented: Bianco et al. Infect Drug Resist. 2018;11:2199-2205, Lindberg et al. Scand J Prim Health Care. 2017 Jun;35(2):178-185 (the study findings showed that more than half of the antibiotic prescriptions were dispensed for the diseases of the respiratory system); Davis et al. Antibiotics 2017; 6,4 23, Bert et al. Eur J Public Health. 2017 Jun 1;27(3):506-512 and Napolitano F et al. PLoS One. 2013 Dec 23;8(12):e84177 to emphasize the pivotal role of patients in reducing the inappropriate and excessive utilization of antibiotics.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #2: Yes: Dr l Grammatico-Guillon, MD, PhD

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

20 Oct 2019

Reviewer #2: The authors took into account the comments of both reviewers and modified the manuscript accordingly.

This improved the response to the objectives of the study and make the article suitable for publication in my opinion.

Response: Many thanks for the positive decision on the manuscript.

Reviewer #3: Many thanks for allowing me to review this revised manuscript.

The authors tried to adequately address the reviewer comments raised in the previous round of review, although they were not able to overcome some shortcomings (e.g. variance estimates, comorbidities and follow-up visits for the same diagnosis were not explored) due to the nature of the study data. However, I believe that inappropriate use of antibiotics is a serious threat to global health, and exploring the pattern of antibiotic prescription among primary care providers represents one of the best tool to promote the appropriateness of use and to control antimicrobial resistance.

The revised manuscript could be approved for publication in PLOS ONE journal, after other minor revision.

Response: Many thanks for your encouraging comments and suggestions. We have incorporated them into our manuscript.

Comment 1: I suggest to add in the limitations section that since the dataset did not include the ICD-10 code, the authors coded themselves the ICD-10 starting from the reason of the visit. It may lead to an inaccurate evaluation.

Response: The suggested limitation has been included in the manuscript in the Discussion section.

Comment 2: Both the reviewers highlighted lack of recent literature on that topic. All these papers have to be cited and commented: Bianco et al. Infect Drug Resist. 2018;11:2199-2205, Lindberg et al. Scand J Prim Health Care. 2017 Jun;35(2):178-185 (the study findings showed that more than half of the antibiotic prescriptions were dispensed for the diseases of the respiratory system); Davis et al. Antibiotics 2017; 6,4 23, Bert et al. Eur J Public Health. 2017 Jun 1;27(3):506-512 and Napolitano F et al. PLoS One. 2013 Dec 23;8(12):e84177 to emphasize the pivotal role of patients in reducing the inappropriate and excessive utilization of antibiotics.

Response: The suggested literature has been incorporated into the discussion section of the manuscript.

Submitted filename: Response to the reviewers.doc

Click here for additional data file.

23 Oct 2019

Outpatient antibiotic prescription rate and pattern in the private sector in India: Evidence from medical audit data

PONE-D-19-14711R2

Dear Dr. Farooqui,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Italo Francesco Angelillo, DDS, MPH

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

5 Nov 2019

PONE-D-19-14711R2

Outpatient antibiotic prescription rate and pattern in the private sector in India: Evidence from medical audit data

Dear Dr. Farooqui:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Professor Italo Francesco Angelillo

Academic Editor

PLOS ONE