Essential medicines for cardiovascular diseases in India: Rapid appraisal of policies and processes at the subnational level

Rapid appraisal of the existing drug procurement and distribution policies of the public sector and processes in Telangana and Madhya Pradesh

We used the rapid appraisal (RA) approach to identify factors affecting variations in state-level access to medicines for CVDs, with a focus on procurement and distribution of essential medicines in two states of India—Madhya Pradesh (low income) and Telangana (high income).

These states were chosen based on convenience to the researchers. The RA approach involves collection and triangulation of multiple data sources to provide an understanding of a situation in a more timely and cost-effective manner than standard social research methods, and to seek a diverse range of perspectives, without aiming for statistical precision.¹³ Triangulation provides internal validity and reliability of the data collected,¹⁴ and these data can then be used to develop specific healthcare interventions.^13,14 RAs have previously been used for understanding health system performance for non-communicable diseases (NCDs) such as diabetes^14–16 and hypertension.^16,17

From March to June 2018, we reviewed publicly available documents related to drug policies, procurement guidelines, EMLs and evaluation reports of the two states. The list of the documents reviewed is given in Supplementary Table I. The documents were obtained from the web portals of drug procurement cells of Telangana and Madhya Pradesh (MP) and peer-reviewed literature databases (including studies regarding the procurement of medication and distribution system at the national and subnational levels) to assess the variations in drug procurement systems in the country. We defined ‘forecasting’ as the process of estimating drug requirements at the facility level for a given period of time (usually one year). Data on indicators for procurement and distribution mechanisms, payment mechanisms and quality control were collected by AG and transferred into a Microsoft Excel spread sheet, and the data were double checked by RS. The template for the document review was based on a previously published study which compared procurement policies of five Indian states.¹² These indicators reflected the procurement cycle: policies, forecasting mechanisms, tendering process, supply chain management, quality control, distribution systems and payment mechanisms.

From April to May 2018, we performed five key informant interviews (three in Telangana and two in MP who were approached through existing networks) of the officers at procurement cells and the staff at the drug warehouses to understand processes that were not in the documents. Our sample includes both the officials from the procurement cells, managing the process from indenting to distribution, and drug warehouse officials, responsible for maintaining stocks and supply of medicines to the clinics. We provided the study information sheet and received verbal consent from officers to carry out the interviews.

The indicators set in the document review and preliminary results from the documents were used to inform the topics for the interview guide in Telangana. The preliminary results from interviews and document review from Telangana informed the interviews in MP. The guide had three sections—general information about EML, current procurement and distribution process and details on essential medicines for CVDs.

The interviews were conducted in English by RS at Hyderabad (Telangana, n=3) and in Hindi by AG at Bhopal (MP, n=2) at the workplaces of participants. RS is a trained qualitative researcher, and AG received training in qualitative interviewing techniques. The interviews in Telangana were audio-recorded, and AG took notes and transcribed the audio recordings. In MP, the interviewees did not give consent for recording. Therefore, AK took notes during the interview. The interview notes in Hindi from Bhopal were translated to English and then back-translated to Hindi by AG.

The ethics committee of the Centre for Chronic Disease Control (Delhi) approved the study.

Comparison of national/subnational essential medicines lists with WHO’s essential medicines list

We searched for national, state (n=29) and Union Territory (n=7) medical services corporation’s online portals of individual states/Union Territories for EMLs between July and September 2018. From the most national and subnational EMLs, we extracted information on the total number of all and essential medicines for CVDs and compared them with WHO’s EML for 2017.¹⁸ We checked whether the EMLs included the following six classes of medicines for CVDs: (i) angiotensin-converting enzyme inhibitors (ACE-I)/angiotensin receptor blockers (ARB); (ii) aspirin; (iii) beta-blockers; (iv) loop or potassium-sparing diuretics; (v) statins; and (vi) thiazide or thiazide-like diuretics as these are the most common medicines for CVDs used in India and globally.¹⁹

We assessed whether the number of classes of essential medicines for CVDs varied by the Empowered Action Group (EAG) status. EAG states are high-focus states for social development, as defined by the Government of India.²⁰ We used Chi-square test at 5% significance to evaluate whether the presence of these six classes of medicines in the EML of the states varied by EAG status using Stata SE 12 software.

Data analysis

We used the framework analysis²¹ to summarize and triangulate data from documentary review and key informant interviews. The study was informed by the WHO framework for access to medicines,² which helped in deriving codes a priori. Broad themes (e.g. general structure and functions of EML and EML committees, procurement, forecasting, supply chain and distribution processes, quality control and assurance procedures and budget) were generated from the framework and further refined using themes identified from the interviews. The themes and the data obtained were written in a descriptive manner and triangulated with the data obtained from document analysis. The qualitative data are not presented as quotes, but the findings have been incorporated within the RA descriptions. Interview recordings and notes were coded using qualitative data software, Open Code v4.02, University of Umeå, Sweden.

Rapid appraisal of existing public sector drug procurement and distribution policies and processes in Telangana and Madhya Pradesh

Telangana, which separated from Andhra Pradesh in June 2014, is a non-EAG state with 31 districts and a population of 35 million.²² The State Health and Family Welfare Department provides health services at the health facilities under the Department of Public Health, Vaidya Vidhana Parishad and Medical Education Department.²³ Telangana State Medical Services Infrastructure and Development Corporation (TSMSIDC), an autonomous organization, is responsible for the procurement and distribution of essential medicines and equipment to the public healthcare facilities.²⁴

MP was formed in 1950 and is an EAG state with 52 districts and a population of 73 million.²⁵ The Directorate of Health Services under the State Health and Family Welfare Department is responsible for all health policies and health services, including drug distribution in the state.²⁶ The MP Public Health Services Corporation Limited (MPPHSCL) procures essential drugs and equipment.²⁷

Essential medicines lists

In both Telangana and MP, the EML is expected to be updated every 2 years (document review).^28,29 The state officials informed during the interview that a new EML committee is formed every 2 years under the leadership of the state health department and consists of experts in various medical specialties. However, EML updates may be delayed due to practical difficulties in convening multiple medical experts (key person interviews). The latest EMLs in Telangana (2015)²⁸ and MP (2018)²⁷ included 557 and 361 medicines, respectively (document review).

EMLs of both states included all six classes of drugs for CVDs. During interviews, we learnt that the EMLs were further stratified as per the level of healthcare facilities to which any particular drug was supplied. Using this unit of analysis, only a subset of drugs for CVDs on both EMLs was available at the primary healthcare level (Table I).

Forecasting, procurement and distribution of essential medicines for CVDs

The information on forecasting, procurement and distribution was obtained during the key person interviews and supplemented or confirmed by documentary reviews.

In both states, essential medicines for CVDs are procured in the general drug procurement system. The funds for medicines from the National Programme for Control of Diabetes, CVDs, Cancers and Stroke (NPCDCS) are pooled along with the state pool of general medicines. The autonomous corporations, TSMSIDC and MPPHSCL, have employed an IT-enabled platform since 2015 for the entire procurement process—from indenting to placing orders—which is known as e-Aushadhi in Telangana²⁴ and MP-Aushadhi in MP.²⁷

Table II compares the pattern of forecasting, procurement and distribution of essential medicines for CVDs in Telangana and MP. The data on the variables are provided by both review of existing state-level documents and key informant interviews. In both the states, the forecasting of annual drug requirements is undertaken before the beginning of the financial year and is based on the consumption from the previous year. In Telangana, each health facility estimates annual requirements, while in MP, forecasting is done at the state level.

When purchase orders are placed by the state corporation (Telangana) or district health officers (MP) for procurement, suppliers send medicines to district warehouses in both states (10 in Telangana and 27 in MP³⁰). Both the states send random samples to the National Accreditation Board for Testing and Calibration Laboratories for external quality testing. The medicines received at the drug stores in both states are quarantined until quality testing is completed before being added to the usable stock.

Healthcare facilities in both states are expected to maintain a 3-month buffer stock of medicines to prevent stock-outs. Healthcare facilities in need of medicines submit their requests to the medical services. A system of diversion of medicines also exists in both states, facilitated by their e-Aushadhi portals, where the medicines from one facility, having unused or excess medicines, can be shifted to another facility in need of those medicines.

In both the states, the total time required from forecasting to drug distribution is estimated to be between 7 and 8 months.

Comparison of Indian national/subnational EMLs with WHO’s EML

The Indian national EML was updated in 2015 (Fig. 1). Among 29 Indian states and 7 Union Territories, we could not identify publicly available EMLs from three states (Goa, Jharkhand and West Bengal) and five Union Territories (Andaman and Nicobar Islands, Chandigarh, Dadra and Nagar Haveli, Daman and Diu and Lakshadweep). The publicly available subnational EMLs were updated between 2012 and 2018. For the states of Arunachal Pradesh, Jammu and Kashmir, Meghalaya and Uttar Pradesh, we could not obtain information on the date of the EML’s latest update.

Close

FIG 1.

Comparison of number of medicines in the publicly available latest essential medicines list of Indian states and Union Territories to Indian national and WHO’s essential medicines list

Export to PPT

The WHO and Indian national EML had 433 and 376 medicines, respectively. The number of medications on state- and Union Territory-level EMLs varied from as low as 153 medicines in Manipur to as high as 1390 medicines in Delhi (Fig. 1).

All six classes of medicines for CVDs were present on the WHO, India’s national and 21 (75%) available EMLs at the subnational level (58% of total states/Union Territories). ACEI/ARBs, beta-blockers and loop or potassium-sparing diuretics were present in all the EMLs at the subnational level. Aspirin was present in 27 (96%) available EMLs, and statins were included in 23 (82%) of the available EMLs (Fig. 2). The selection of essential medicines for CVDs into the EML did not differ significantly by EAG status (p=0.45). Many northeastern states (e.g. Arunachal Pradesh, Bihar, Mizoram, Nagaland and Sikkim) did not have statins on their EML (Table III). Statins, thiazide and thiazide-like diuretics were absent in Bihar and Arunachal Pradesh. At the same time, Arunachal Pradesh also did not include aspirin in addition to statins and thiazide and thiazide-like diuretics.

Close

FIG 2.

Proportion of Indian states and Union Territories having cardiovascular disease medicine classes in essential medicines list; Comparison across Empowered Action Groups versus Non-Empowered Action Groups

Export to PPT

Strengths and limitations

Our study is one of the first to document the variability of the selection of medicines for CVDs at the state level and reporting an in-depth situation analysis of the procurement systems using predetermined indicators providing a macro-level evaluation. Another strength of the study is the use of the RA approach to triangulate data from multiple sources (document review and qualitative methods) providing validity and reliability of the data collected. The study has limitations: first, data on the EMLs at the subnational level were captured only from the publicly available portals; second, data on outcome indicators, such as the availability of medicines at the point of purchase by patients, were not collected. However, our upstream evaluation can help interpret the results of future studies that capture such data. Our work also suggests that future research efforts should be focused on studying the drug procurement processes at the subnational level and its implications on the availability, accessibility and pricing of essential medicines. Future efforts should also focus on strengthening EMLs at the subnational level and the selection process to match disease burden for rational and evidence-informed decision-making. The management of CVDs is also dependent on the screening and diagnostic capacities of the health system. Finally, in addition to the EML, the WHO has created essential diagnostics lists, which include diagnostic tools such as serum cholesterol monitoring and blood pressure measurement devices.⁴⁴ These tools were not the focus of this research study and can be pursued in future research efforts.

Conclusion

India is one of the largest producers of generic medicines in the world. All six classes of essential drugs for CVDs are manufactured in the generic version in India, and state-level public health departments can potentially procure them at relatively low costs. Addressing the challenges that we identified in our study, namely listing on EMLs at the subnational level, appropriate forecasting and reducing lead time in procurement may help to ensure that essential medicines required to prevent, treat and control CVDs are available. Our study provides an in-depth analysis of how policies and processes for medication procurement at the subnational level.