Perceptions of effective policy interventions and strategies to address antibiotic misuse within primary healthcare in India: A qualitative study

Abstract

Objectives

Antimicrobial resistance poses a major public health threat. Despite Indian retail sector antibiotic consumption per capita increasing by approximately 22% between 2008 and 2016, empirical studies that examine policy or behavioural interventions addressing antibiotic misuse in primary healthcare are scarce. Our study aimed to assess perceptions of interventions and gaps in policy and practice with respect to outpatient antibiotic misuse in India.

Methods

We conducted 23 semi-structured, in-depth interviews with a variety of key informants with diverse backgrounds in academia, non-government organisations, policy, advocacy, pharmacy, medicine and others. Data were charted into a framework matrix and analysed using a hybrid, inductive and deductive thematic analysis. Themes were analysed and organised according to the socio-ecological model at various levels ranging from the individual to the enabling environment.

Results

Key informants largely focused on the importance of adopting a structural perspective to addressing socio-ecological drivers of antibiotic misuse. There was a recognition that educational interventions targeting individual or interpersonal interactions were largely ineffective, and policy interventions should incorporate behavioural nudge interventions, improve the healthcare infrastructure and embrace task shifting to rectify staffing disparities in rural areas.

Conclusions

Prescription behaviour is perceived to be governed by structural issues of access and limitations in public health infrastructure that create an enabling environment for antibiotic overuse. Interventions should move beyond a clinical and individual focus on behaviour change with respect to antimicrobial resistance and aim for structural alignment between existing disease specific programs and between the informal and formal sector of healthcare delivery in India.

Background

The World Health Organisation (WHO) has declared antimicrobial resistance (AMR) as one of the top ten global health threats (WHO, 2020; Miller-Petrie and Gelband, 2017). Much of the current literature on addressing AMR in LMICs has focused on antibiotic stewardship, defined as interventions that seek to measure and optimise the use of antibiotics; however, the predominant focus is on inpatient settings, where antibiotic stewardship committees can be readily established and compliance can be monitored (Van Dijck et al., 2018; Davey et al., 2013; Baubie et al., 2019; Schuts et al., 2016; McKnight et al., 2019; Kommalur et al., 2021; Bhalla et al., 2016).

However, outpatient antibiotic use is a significant driver of resistance, and antibiotic consumption per capita in the retail sector in India increased by approximately 22% between 2008 to 2016 (Farooqui et al., 2018). Despite this, there are few evidence-based interventions to address antimicrobial misuse in primary healthcare or outpatient settings in LMICs (Brinkmann et al., 2020; Ivanovska et al., 2013; Senn et al., 2014; Bhavnani et al., 2007; Gautham and Shyamprasad, 2010).

In the Indian context, antibiotic use in primary healthcare is difficult to contextualise without a concrete understanding of the decentralised and pluralistic model of healthcare delivery in rural and urban settings. India has a long-standing tradition of multi-tiered, pluralistic care seeking behaviour. Outpatient care is provided free-of-cost at government facilities, and India’s decentralised health system aims to provide one primary health centre (PHC) for every 30,000 people and one community health centre (CHC) for every 120,000 people (Directorate of Health Services, 2022).

In reality, there are far fewer facilities and shortages of doctors and other qualified providers in rural areas (Gautham and Shyamprasad, 2010; National Health Mission, 2020). Such limitations in the public sector often force patients to seek care in the private sector, access antibiotics over-the-counter at pharmacies as ‘quick cures’ to alleviate symptoms, or visit informal providers in rural areas (Kumar et al., 2007; Das et al., 2016; Gautham et al., 2014; Gupta and Patel, 2021; Nair et al., 2019).

This study focuses specifically on outpatient antibiotic misuse and explores the following research questions:

• To what extent have current interventions to address outpatient antibiotic misuse been successful?

• What are the current gaps and opportunities in policy and practice with respect to antibiotic misuse?

Methods

Design and data collection

Semi-structured, in-depth interviews were conducted with 23 key informants, including policy makers, non-profit leaders working in AMR, researchers, academics, allopathic medical practitioners, public health practitioners, among others. Through a literature review, we identified and contacted a list of key informant experts who worked in the area of antibiotic misuse in India. We conducted a detailed review of their publications, prior research on antibiotic use and a digital review of their engagements with other AMR stakeholders (e.g. Twitter and social media presence). Participants were usually senior academics who also had medical or public health training: some were affiliated with academic institutions like the London School of Hygiene and Tropical Medicine, while others were senior academics at local AMR-focused organisations such as the Centre for Disease Dynamics, Economics and Policy or the Central Drugs Standard Control Organisation. Subsequent snowball sampling was also utilised to identify and recruit additional participants after an initial round of outreach. Participants were continuously and purposively sampled until saturation was attained.

Given their subject matter expertise, this study analysed key informant perceptions of policy strategies and interventions to outpatient antibiotic misuse in India. Each interview was transcribed verbatim and verified against the audio recording for transcription accuracy. Identifying information and names were redacted from each transcript. All interviews were conducted over Zoom in English, and no translation was required.

Data analysis

Data were analysed using a framework matrix methodology, which incorporates flexible and robust approaches from thematic analyses and qualitative content analyses (Gale et al., 2013). Each transcript was read and re-read to aid familiarisation with the data. The codebook was developed iteratively, using both an inductive and deductive approach: an initial set of codes were derived based on a review of the literature, interview guide and familiarisation with the first few transcripts, while emergent codes were added based on a review of each subsequent transcript. The framework matrix codebook has been included as Supplementary Material 1.

Once all codes were saturated in the matrix, a cross-case comparison for each code was completed, detailed notes were referenced during the development of the codebook and emergent thematic areas from the data were identified. The matrix allowed for a systematic comparison by case and code, enabling an analysis of higher-level themes across the entire dataset.

A social-ecological model was utilised to better organise themes and present interventions related to antibiotic use at various interconnected levels: individual, community and policy/enabling environment (The Borgen Project, 2017; CDC 2021). The social-ecological model is widely used in public health research and practice as it underscores the complex interplay between a range of factors.

Ethics

The study was approved by the Maastricht University Ethics Review Board (FHML-REC/2020/044). All interviewees were provided with an information sheet, agreed to be audio recorded, and each participant provided written informed consent by e-mail prior to participation in the interview.

Results

Semi-structured interviews were conducted with 23 key informants in total who represented a range of perspectives in academia, medicine, pharmacy, policy making and non-profit advocacy within India. Most respondents were part of academic institutions (38%) and several overlapped with government/policy making work (18%) and medicine (23%). 9% identified as pharmacists and 12% worked in the non-profit or advocacy sector.

61% of the respondents were female, and 39% were male; respondents ranged in age from 29 to 63 years old. The themes were analysed and charted according to the social-ecological model and framework:

Individual or interpersonal level interventions

Limited impact of educational or training interventions

Interventions that targeted individual or interpersonal interactions were perceived to inadequately recognise the context and influence of societal or systemic factors on behaviour and were considered ineffective according to most key informants. Economic incentives and a desire to maximise profit under a capitalistic, market-driven model of healthcare delivery, were perceived as major drivers of antibiotic use: ‘It's not a knowledge deficit that’s happening here…for private providers especially, it’s to their own economic detriment to sort of mitigate against overuse. So I think it’s a very difficult thing to ask private providers to put their livelihood on the line’ (KII 21).

One-off educational or training interventions targeting the individual were perceived as unlikely to produce significant behaviour change: ‘[we have found that] training improved correct management of the cases, but it didn’t decrease unnecessary usage of medicines…doctors who are better tend to do the right thing more often, but they don’t do the wrong thing less often’ (KII 22).

Multi-pronged interventions, which combine knowledge enhancement with stronger regulations, audit and feedback or peer-to-peer comparison, were perceived to be more effective in creating behaviour change: ‘I think it has to be a multi-pronged approach. And it has to be something that is continuously reinforced. But behavioural change is very tough’ (KII 20).

Community-level interventions

Interventions that went beyond individual or institutional actors and analysed the networks between these actors within the context of the broader community were classified as community-level interventions.

Adoption of hub-and-spoke models for outreach with informal providers

Key informants noted that informal providers already have strong collaborations and referral networks with formal providers, which can be leveraged for future interventions. They are extremely well-organised and have associations that coordinate with one another via WhatsApp groups: ‘these guys are way more powerful and they’re better organised [than] we think…[you could] try to convince one to let you in on his WhatsApp group’ (KII 18).

Pharmaceutical companies have already tapped into these associations to increase sales of antibiotics in these unsaturated markets. These connections were often described as a hub-and-spoke approach to building educational networks, and were perceived to be helpful touch points for future interventions:

‘How do you work with multiple stakeholders who have such diverse interests, not just across the human and veterinary health sectors, but also in terms of economic orientation, in terms of their professional background? We are dealing with informal providers, we’re dealing with formal doctors, we realised that you can’t just address informal providers, without bringing in doctors- the links between them are too strong and too close. So, you know, we need to address them as...a kind of hub-and-spoke...because that's how they also function’ (KII 7).

Leveraging task shifting interventions

Key informants perceived that access group antibiotics that have lower potential for resistance should be widely available, watch group antibiotics should be more limited given their higher potential for resistance and reserve group antibiotics should only be used as a last resort. Auxiliary nurse midwives (ANMs), who work in PHCs, were perceived by a few key informants as a crucial pillar in gating access to these antibiotics:

‘It’s probably important when you look at the access, watch, and reserve list of antibiotics, to ensure that the access antibiotics are available…the ANM for example…know the villages, they know exactly what's going on…probably the government will have to have some kind of a primary list of access antibiotics based on certain science and protocols’. (KII 13).

This kind of task shifting intervention could integrate informal providers as well: ‘[policy makers] haven’t tapped on to that sector to provide access to health care...so for example, for a common cold or cough or injury, it’s fine to go to this kind of guy who can provide you first aid access, as long as he’s trained, he’s certified’. (KII 14).

Key informants noted that this approach could limit access to watch or reserve group antibiotics, empower frontline providers who already have strong connections in the community and conserve important antibiotics without hampering access to necessary medicines for the community.

Policy/enabling level interventions

Key informants predominantly named systemic, policy-level interventions as promising avenues for curbing antibiotic misuse and enacting behaviour change, including drug regulatory policies and market-based incentives for the retail and pharmaceutical sector.

Leveraging the role of pharmaceutical companies and representatives

The role and influence of pharmaceutical company representatives varies widely across outpatient settings in different Indian states. As one key informant stated, while it is true that the nexus between the pharmaceutical industry and the medical profession ‘begets more inappropriate prescribing’, PCRs occasionally function as ‘medical science liaisons or medical advisors’ who engage doctors in a ‘constant adaptation of knowledge of the recent data’ (KII 5). Participants mentioned that PCRs have a fixed salary component and a variable component that increases based on the volume of sales, which can incentivise heavy marketing and over-prescription of antibiotics. However, several key informants who were also medical doctors reported that these interactions with PCRs also allowed them to learn about new drugs on the market.

Key informants suggested working directly with pharmaceutical companies who have previously made pledges to delink sales from incentives for employees. ‘In 2019, I think there were many pharma companies that decided to delink their sales with the profit, right?...But eventually, [they] went back on [their] own promises to delink sales with profit for the employees, because no other company was following suit’ (KII 14). This approach was largely recognised as an uphill battle as it was at odds with the market-driven, biomedical model, but there was some optimism for companies to follow suit and rebrand their healthcare approaches around conservation of antibiotics.

There remain significant challenges to large-scale changes in pharmaceutical regulation in India. As some key informants mentioned, the pharmaceutical industry has deepened its presence in ‘unsaturated’ rural and informal markets, which ‘hold huge potential for expanding revenue segments’ (KII 7). Decoupling volume of sales from the salary would fundamentally alter the incentives for marketing and is unlikely to happen without ‘very strong regulatory provisions’ (KII 12).

Inconsistent implementation of drug regulatory policies (e.g. Schedule H1)

India has strong regulations on paper that limit the prescriptions and sales of antibiotics, but key informants stated that implementation and enforcement have consistently lagged behind. ‘If you think about the over-the-counter sale of antibiotics, this is technically forbidden in India, but it happens all the time...throughout drug stores at every corner’ (KII 1).

Schedule H1, an amendment to the Drugs and Cosmetics Rules Act of 1945 that imposed restrictions on over-the-counter dispensing of certain antibiotics (mostly third and fourth generation cephalosporins, carbapenems and newer fluoroquinolones), was perceived as a vital policy with ineffective implementation and adherence. Under the policy, drug inspectors have the authority to conduct surprise checks, but these are not regularly conducted, according to a senior authority from the Indian Pharmaceutical Association: ‘schedule H1 [is]...checked sometimes by the drug inspectors...but maintenance of the Schedule H1 register only [entails] writing down the prescription regimen. Beyond that, [I have] not seen any [checks]’ (KII 11).

Poor enforcement and implementation of regulatory policies also impacts the manufacturing pipeline for antibiotics and the types of formulations which are available to the public, such as fixed-dose combinations: ‘Why don’t we curb the manufacturing of antibiotics? And why don’t we become more serious about the FDCs [fixed-dose combinations] in the case of antibiotics?...look at the number of formulations which are available in India…the regulators are always ill equipped with the infrastructure, the number of drug inspectors’ (KII 11).

Balancing access to antibiotics while limiting excess antibiotic use

Several key informants noted that interventions must recognise the role of systemic and structural problems, such as inadequate hygiene and sanitation measures, poor public health infrastructure, inadequate access to antibiotics and limited frontline healthcare staff (especially in rural areas), in creating endemic and problematic access barriers within the Indian healthcare system. ‘Irrational’ antibiotic use in this regard was understood as a strategic response by healthcare providers to cope with inadequate access to healthcare for the vast majority of the Indian population. From this perspective, the term ‘irrational’ was seen as a misnomer:

‘I think the white world calls it irrational practice. But how can you call it irrational when there are clearly good reasons for something that's taking place?’ (KII 4).

‘Then, of course, in places like India, and any low-and-middle-income countries, as always, an issue of access and excess. So, you got rural places where there's hardly any healthcare facilities really...And if there is no pharmacy shop, what do you do? Then you go to a quack, right? Or you go to a native medicine person. And so all these are structural challenges, which actually prevent or rather inhibit forward movement in terms of action plans’. (KII 13)

These social determinants of antibiotic prescription behaviour were seen as critical levers in understanding why healthcare providers dole out seemingly ‘irrational’ prescriptions. There was also a concern that policy interventions must be cognisant of unintended consequences that could result from any policies that restrict access to antibiotics in the general population in the absence of a strong and functioning healthcare system.

Discussion

The importance of recognising and addressing structural barriers to accessing high quality healthcare and understanding antibiotic use in the context of social determinants of health was a clear overarching theme across our interviews. Our results indicate that prescription behaviour is perceived to be governed by structural issues of access and limitations in public health infrastructure that create an enabling environment for antibiotic overuse. Thus, from a socio-ecological perspective, interventions designed to improve knowledge of antibiotic use or educational interventions targeting individual or interpersonal interactions alone are unlikely to work, as there is a clear knowledge-practice gap even with interventions that successfully improve an individual’s existing knowledge base.

This gap between knowledge and practice is well established in the global literature in AMR. For instance, Pearson and Chandler conducted more than 200 qualitative interviews and field observations in seven study sites across Ethiopia, India, Nigeria, the Philippines, Sierra Leone and Vietnam (Pearson and Chandler, 2019). Their study found that even when awareness of AMR was high among human and animal healthcare professionals, it did not translate into better practice and reduced prescribing. Contextual factors, such as improved infrastructure and regulation, were far more salient in reducing reliance on antibiotics (Pearson and Chandler, 2019).

In India, the formal leadership of the Indian Medical Association advocates for a strict ban on antibiotic dispensing by informal providers, along with regulatory enforcement and punishment, in accordance with the provisions of the Indian Medical Council (IMC) Act of 1965, which makes it illegal for anyone without a university medical qualification to practice modern medicine (Chowdhury, Gautham, and Kumar, 2018; Davis, 2016; Press Information Bureau, 2022).

Our key informant interviews stated that wishing these frontline providers away in the context of lacking political will is both impractical and detrimental to public health. They suggested a newer approach: to empower formal and informal frontline providers, train and equip them with access group antibiotics, leverage existing connections and referral networks to private or public formal providers and explicitly regulate access to watch or reserve group antibiotics.

In addition to a strengthened frontline healthcare workforce, additional investments must be made in implementing a routine prescription audit and feedback system in government PHCs, and building partnerships with academic and community-based organisations to establish hub-and-spoke models via associations of formal and informal providers, which could enable a thorough mapping of informal providers operating in the region.

Several studies have examined the behaviour of private practitioners who treat tuberculosis, and the divergence in behaviours in terms of referrals to government hospitals (Ecks and Harper, 2013; Brhlikova, 2011; Engel and Van Lente, 2014; McDowell and Pai, 2016; Yellappa et al., 2018). McDowell and Pai found that even non-biomedical, AYUSH private practitioners frequently resorted to biomedical therapies, and almost 94% referred patients to a chest physician or the public sector (McDowell and Pai, 2016). Engel and van Lente found that effective integration of a public-private mix requires bridging diverse organisational cultures: the command-and-control style of the Revised National Tuberculosis Control Program, the professional monopoly of private practitioners that is aversive to control and the community-oriented and participatory model of NGOs (Engel and Van Lente, 2014). Many lessons for controlling AMR can be gleaned from tuberculosis treatment efforts in India in this regard, given the motley mix of public providers, private practitioners and traditional healers in the TB care continuum.

Finally, the pharmaceutical industry must be strategically engaged to counter the incentives for overprescribing, which are inherently built into the business model. Practices that leverage a firm’s desire for responsible stewardship-focused branding might be useful to engage pharmaceutical companies that have previously expressed interest in acting against AMR. The overall regulation of the pharmaceutical industry is another major policy area that requires additional oversight and intervention.

While the Government has enacted some regulations with respect to banning FDCs and enacting restrictive policies such as Schedule H1, their success has been a mixed bag with poor enforcement mechanisms and inadequate implementation across the country (Press Information Bureau, 2022). A 2018 study found that more than 60% of FDCs actually had no regulatory approval and many of the formulations were pharmacologically incompatible, reinforcing the need for better regulation and oversight of the pharmaceutical industry (McGettigan et al., 2019). Regulatory recommendations inevitably face bureaucratic hurdles in implementation in a country as diverse and decentralised as India, but the Government has been able to implement and strictly enforce restrictions in areas it has deemed important, such as opioid access (Cleary et al., 2013). However, these have often created major access-related barriers for patients, and great care must be taken to ensure that patient rights to medicine are taken into account and existing disparities in access are not worsened.

The prevailing clinical focus on individual prescriber behaviour in AMR must consider the structural determinants that govern behaviour from a socio-ecological perspective. As Broom et al. detail in their examination of AMR from three different continents, an effective and sustainable approach to tackling AMR requires ‘solidaristic models that espouse collective responsibility and recognise relative opportunity to act rather than a continuation of the individualistic behavioural models that have, so far, proven largely ineffective’ (Broom et al., 2021).

Limitations

While this study addressed a critical gap in the literature and incorporated unique insights from key informants, both academics and community-based advocates alike, it was not without limitations. Respondent validation through informal interviews with participants after data collection and analysis was not possible due to time and resource limitations. Even though 53 key informants were invited to participate, 57% declined the invitation to participate largely due to time constraints and non-response during the height of the COVID-19 pandemic; as a result, there were more perspectives from academia and medicine or public health compared to government institutions and non-allopathic providers.

Conclusions

Behaviour change interventions are very difficult to sustain without addressing structural barriers to accessing care that influence the policy and enabling environment. The healthcare delivery model within India offers unique opportunities to address structural barriers to access care and strengthen the existing delivery of primary and outpatient primary healthcare, while leveraging traditional and informal healthcare practitioners as a frontline workforce and improving regulatory efforts within the pharmaceutical industry.

Supplemental Material

Supplemental Material - Perceptions of effective policy interventions and strategies to address antibiotic misuse within primary healthcare in India: A qualitative study

Supplemental Material for Perceptions of effective policy interventions and strategies to address antibiotic misuse within primary healthcare in India: A qualitative study by Mohit Nair, Nora Enge, Maurice P Zeegers and Sakib Burza in Journal of Infection Prevention