Nan Zhou  ( Department of Pharmacy, Shaanxi Provincial People’s Hospital, China. )
Zhiyuan Fang  ( Department of Pharmacy, Shaanxi Provincial People’s Hospital, China. )
Dong Liang  ( Department of Information, Shaanxi Provincial People’s Hospital, China. )
Jiangping Lian  ( Department of Pharmacy, Shaanxi Provincial People’s Hospital, China. )
Huan Han  ( Department of Pharmacy, Shaanxi Provincial People’s Hospital, China. )
Rui Li  ( Department of Neurology, Shaanxi Provincial People’s Hospital, China. )

Objectives: To evaluate the effect of a pharmacovigilance system on potentially inappropriate medication prescriptions for elderly patients.

 

Method: The retrospective study was conducted at Shaanxi Provincial People’s Hospital, China, after approval from the ethics review committee, and comprised data from May 2020 to April 2021, and comprised prescriptions related to elderly patients aged at least 65 years. Number of medication risk assessment entries, number of intervened medical orders on outpatients and inpatients number of medical order prompts, and number of physician communication with prescription-checking pharmacists were noted. Potential drug interaction rate was compared between pre- implementation from May to October 2020 and post-implementation from November 2020 to April 2021. Besides, the usage of sedatives and hypnotics and potentially inappropriate medication was noted for the period from January to June 2021 to evaluate the sustained effect of pharmacovigilance system. Data was analysed using SPSS 19.

 

Results: A total of 118 drugs were involved in the 3911 entries of outpatient prescription warnings, of which 19 drugs accounted for 3156 (80%). Besides, a total of 113 drugs were involved in the 3999 entries of inpatient prescription warnings, of which 19 drugs accounted for 3199 (80%) The overall prevalence of potentially inappropriate medication related to sedatives and hypnotics decreased post-intervention as warning percentage was 16.1% in January and 6.7% in June among outpatients. On inpatients, the warning percentage was 30.6% in January and 6.1% in June.

 

Conclusions: The pharmacovigilance system could reduce potentially inappropriate medication and provide deeper technical support for the safety of medical behaviour and individualised treatment of patients.

 

Key Words: Pharmacovigilance, Drug safety, Potentially inappropriate medication, Elderly patients, Risk management.

 

(JPMA 73: 525; 2023) DOI: 10.47391/JPMA.6428

 

Submission completion date: 17-03-2022 — Acceptance date: 24-09-2022

 

Introduction

 

With rapid development and improvement of living standards, the average life expectancy of humans has increased. The ageing of the population has become a serious problem faced by all countries. Multiple diseases are often found in an elderly person concurrently, and most are chronic. A wide variety of drugs are available for treatment, but have prominent drugs interaction problems. Additionally, the pharmacokinetics in the elderly population is quite different from that of other age groups. Therefore, drugs easily accumulate in the body and cause adverse reactions¹.

In 1991, the criteria for determining inappropriate medication use in sanatorium residents were published for the first time by American geriatric experts², known as the Beers criteria. Subsequently, the Beers criteria have been updated several times by the American Geriatrics Society (AGS) following a comprehensive and systematic evaluation and analysis of relevant studies^3-5. A list of drugs that should be avoided in elderly patients as such, and under specific disease conditions, drugs to be used with caution in elderly patients, marking of evidence level and strength of recommendation for each item was established. To date, the criteria for potentially inappropriate medication (PIM) use in older adults applicable to each local area have been formulated by many countries based on the Beers criteria to monitor and guide rational medication usage among the elderly^6-11. However, it is unrealistic to assume that the list of PIMs in older adults and specific risk contents are remembered by every physician, nurse practitioner or pharmacist. This highlights the need for processes to support prescribers in making rational choices, such as a software that may issue warnings when PIMs are prescribed. A pharmacovigilance system (PVS) was established in Shaanxi Provincial People’s Hospital to issue PIM alerts. The current study was planned to evaluate the effect of the PVS on PIM prescriptions for the elderly in a tertiary care setting.

 

Materials and Methods

 

The retrospective study was conducted at Shaanxi Provincial People’s Hospital, China, after approval from the ethics review committee and comprised data from May 2020 to April 2021.

International and domestic guidelines and consensus documents^3-5 on medication use for elderly patients were identified to establish PVS rules. The content was embedded in the medical order running software. The real-time audit of different dimensions was conducted on the issued medical orders, and the medical orders were intercepted or prompted on the hospital information system (HIS) to ensure medication safety.

The contents were updated in real time in line with evidence-based medicine (EVM) International and domestic guidelines and consensus documents published by the National Centre for Adverse Drug Reaction Monitoring¹². The PVS was intended for use in adults 65 years and older in all ambulatory, acute, and institutionalised settings of care. Hospice and palliative care settings were excluded. Five types of risk assessment are conducted: PIM related to diseases or symptoms in older adults; medications used cautiously in older adults; drugs interactions that should be avoided in older adults; drugs that should be avoided; an drugs that should be reduced in dose in patients with renal insufficiency.

Each risk assessment rule was collated with EVM, and source of recommendation was marked. The risk assessment rule entries were classified and summarised. By comparing the list of risky drugs with the list of drugs used in the hospital, the list of drugs that posed risks to elderly patients and risk assessment rules for the hospital were screened to establish and complete the document information.

The risk assessment rules were imported into the HIS in batches (Figure).

 

 

The real time audit of different dimensions was conducted on the medical orders issued according to the different contents of the entries in the PVS. In case of risks, early warning was conducted, with warning levels divided into three situations: interception, prompting, or online communication with prescription-checking pharmacist. For drugs that are prohibited for older adults or the combination of drugs with serious interactions or where the drugs have a significant impact on the original disease, such a medical order was directly intercepted by the system and a dialogue box explained the reasons for interception and evidence. In view of the situation that older adults use drugs with caution or require drug dose reduction or in case of drug interactions with low potential risks, only the prompt dialogue box was shown by the system, explaining the reason and EVM. The level of risk was determined by physicians according to the specific situation, and whether or not to modify the medical order was determined by them. In cases where the physician had objections or doubts on the warning content, communication was conducted online with prescription-checking pharmacists to change the disposition of the medical orders.

In the current study, prescriptions related to elderly patients aged at least 65 years were evaluated. Number of medication risk assessment entries, number of intervened medical orders on outpatients and inpatients, number of medical order prompts, and number of physician communication with prescription-checking pharmacists were noted. Potential drug interaction rate was compared between pre-implementation phase from May to October 2020 and post-implementation phase from November 2020 to April 2021. Besides, the usage of sedatives and hypnotics, like estazolam, clonazepam, diazepam, zolpidem and alprazolam, and PIM was noted for the period from January to June 2021 to evaluate the sustained effect of pharmacovigilance system.

 Data was analysed using SPSS 19. Data was presented as mean ± standard error of the mean (SEM). Differences between groups were analysed using paired samples t-test. The level of statistical significance was set at p<0.05.

 

Results

 

A total of 343 drugs posing risk to the elderly were classified, involving 1164 warnings. Through cross-comparison between the list of risky drugs and list of drugs used in the hospital, a total of 418 drugs, including generics of an original drug, were screened, and 10,663 audit rules were summarised.

The PVS was implemented on a trial basis in the Neurology, Respiratory, Cardiology, Nephrology and Gastroenterology departments. The PVS provided 3911 warnings related to 1767 outpatients (10.5% of the total number of outpatients). Among these, prescriptions were intercepted 349 times. Communications between physicians and pharmacists were 138 times. Finally, 263 medical orders were changed by physicians.

Among the inpatients, the PVS provided warnings 3999 times and 844 patients (23.7% of the total number of inpatients) were involved. Among these, the prescriptions were intercepted 230 times. Communications between physicians and pharmacists were 120 times. Finally, 146 medical orders were changed by physicians.

A total of 118 drugs were involved in the 3911 entries of outpatient prescription warning, of which 19 drugs accounted for 3156 (80%) (Table 1),

 

 

and most of them were psychotropic, cardiovascular as well as cardio and cerebrovascular drugs. A total of 3510 (89.7%) warning levels were mainly prompts. Drugs that were intercepted included three sedatives and hypnotics (estazolam, clonazepam, diazepam), one nonsteroidal anti-inflammatory drug (NSAID) (oxaprazine) and one psychotropic drug (quetiapine). PVS had great influence on the usage of sedatives and hypnotics (estazolam, clonazepam, diazepam, zolpidem, alprazolam). The usage of the five drugs decreased significantly over the January-June period. Number of prescriptions of the five drugs were 38 (0.5% of total number of prescriptions), 35 (0.5%), 40 (0.6%), 41(0.6%) and 152 (2.2%) respectively in January 2021. The proportion decreased to 21 (0.4%), 15 (0.3%), 17 (0.3%), 20 (0.4%) and 103 (2%) respectively in June 2021 (t=6.325, p=0.003).

A total of 113 drugs were involved in the 3999 entries of inpatient prescription warning, of which 19 drugs accounted for 3199 (80%) (Table 2),

 

 

and most of them were psychotropic and cardiovascular drugs. 3733 (93.3%) warning levels were mainly prompts.

The overall prevalence of PIM decreased during the 6 months. On outpatients, warning percentage was 16.1% in January 2021 and 6.7% in June 2021. On inpatients, warning percentage was 30.6% in January 2021 and 6.1% in June 2021. Besides, the potential drug interaction rate was also reduced. Interaction was 83 (2.7% of the total number of inpatients) on inpatients and 186 (0.8% of the total number of outpatients) on outpatients during pre-intervention. Interaction decreased to 72 (2% of the total number of inpatients) on inpatients and 166 (0.6% of the total number of outpatients) on outpatients during post-intervention.

 

Discussion

 

According to the World Health Organisation (WHO), the ageing population will exceed 1.5 billion in 2050, accounting for 16% of the global population. As shown in the National Annual Report on Adverse Drug Reaction (ADR) Monitoring (2018), of the 1.499 million reports received, elderly patients aged >65 years accounted for 27.7%¹³. Such patients have a high risk of adverse drug events^14-15. Therefore, the implementation of PVS contributes to prompting clinicians and pharmacists of the medication risk in real time. More attention should be given when issuing or reviewing medical orders so as to reduce the occurrence of drug injury events, and to play a role in popularising health sciences for the public^16-17.

In the present study, the drugs with the highest incidence of early warning were anti-insomnia drugs, including zolpidem, alprazolam, diazepam and estazolam. It is found that >50% of the elderly have insomnia, and 23% of them have severe insomnia¹⁸, which may be the reason for the high frequency of use of such drugs in elderly patients. During the rule formulation, the warning level of anti-insomnia drugs with high risk and long half-life was set as interception, whereas drugs with short and medium half-life were still maintained as prompts, considering that the extremely high warning frequency would easily lead to ‘vigilance fatigue’. Taking diazepam as an example, the long half-life can easily lead to a ‘hangover phenomenon’ in the early morning after using the drug in elderly patients, that is, drowsiness, dizziness, sleepiness and disorientation, which is similar to the symptoms of drunkenness. The present study revealed that accidental injuries in the elderly population, such as falls and fractures, are associated with long-term use of long-acting sleeping pills, such as diazepam^19-20. It is recommended that physicians must analyse the cause of insomnia before using the drug. First, psychological or behavioural therapy should be considered, or short-acting hypnotics should be used for treatment²¹, PVS indeed had great influence on the usage of sedatives and hypnotics.

Antiplatelet and anticoagulant drugs are second only to anti-insomnia drugs. The high incidence of cardiovascular and cerebrovascular diseases contributes to the frequent use of these drugs²². For secondary prophylaxis with definite evidence, the warning level is prompt. Proton pump inhibitors (PPIs) are commonly used by older adults. ADR caused by long-term treatment have been recorded in numerous studies²³, such as induced infection, bone loss and fracture risk. However, this situation is usually overlooked in treatment. Prompts are provided by the PVS for such problems.

The vast majority of elderly patients have chronic pain, such as arthritis, gout and other diseases. NSAIDs are still the first choice for pain relief²⁴, indirectly resulting in an increase in the number of individuals with peptic ulcers. The older the patient, the worse the physiological functions of the body and the weaker are liver and kidney functions1. The ability of prostaglandin synthesis is weakened, which leads to a decrease in gastrointestinal mucosal defence function, and NSAIDs-related peptic ulcer is more likely to develop²⁵. Prompts are provided by the PVS for the course of NSAIDs, thereby reducing the risk factors.

The use of two or more centrally acting drugs counts for most of the early warning of drug interaction. Two sedative and hypnotic drugs were used alternately. Antidepressants were used in combination with sedative and hypnotic drugs. It might increase the central adverse reactions in elderly patients. Fortunately, after PVS warning, the potential drug interaction rate hugely decreased. It reduced the potential risk of ADRs.

Furthermore, the early warning frequency of insulin was also high in the present study. Elderly patients have relatively large blood glucose fluctuations and high probability of hypoglycaemia because of their special fitness²⁶. Insulin is commonly used in elderly patients with diabetes. Great progress has been made in insulin-related research and development, including continuous optimisation of pharmacokinetics or administrative methods, but ADRs, such as allergic reactions, hypoglycaemia and weight gain, are still inevitable. In case of severe hypoglycaemia, consciousness will be lost and even life will be endangered²⁷. Therefore, insulin should be used with caution in older adults. If insulin is inevitably used, medication education and blood glucose monitoring for patients should be strengthened²⁸. Therefore, early warning of insulin is indispensable.

Elderly patients were still at risk although PIM prescriptions reduced. Most (60%-70%) pharmacist interventions were accepted. Although these interventions were performed in some cases after ignoring the initial alert, pharmacist interventions are frequently ignored. The benefits of PVS were ease of access and the provision of specific warnings to physicians when prescribing PIMs. However, polypharmacy has been continuously prescribed in elderly patients that is associated with a higher prevalence of PIM. These issues should be addressed by de-prescribing to minimise prescription of multiple medicines and to withdraw PIM in elderly patients^29-30.

In terms of limitations, the current study was conducted at a single hospital, and the results may not be applicable to other hospitals due to their unique PVS alert requirements. Moreover, because of limited data access, observations were conducted only in the Neurology, Respiratory, Cardiology, Nephrology and Gastroenterology departments. Therefore, findings may not be generalisable.

 

Conclusions

The PVS could reduce PIM and provide deeper technical support for the safety of medical behaviour and individualised treatment of patients. With the establishment of PVS, inappropriate prescription can be effectively reduced.

 

Disclaimer: None.

 

Conflict of Interest: None.

 

Source of Funding: National Natural Science Foundation of China (grant nos. 81703514), Shaanxi New-star Plan of Science and Technology (grant nos. 2021KJXX-23) and National Natural Science Foundation of Shaanxi Provincial (grant nos. 2020JQ-946).

 

References

 

1.      Alagiakrishnan K, Ballermann M, Rolfson D, Mohindra K, Sadowski CA, Ausford A, et al. Utilization of computerized clinical decision support for potentially inappropriate medications. Clin Interv Aging. 2019; 14:753-62. doi: 10.2147/CIA.S192927.

2.      Beers MH, Ouslander JG, Rollingher I, Reuben DB, Brooks J, Beck JC. Explicit criteria for determining inappropriate medication use in nursing home residents. UCLA Division of Geriatric Medicine. Arch Intern Med. 1991; 151:1825-32.doi: 10.1001/archinte.1991.00400090107019.

3.      American Geriatrics Society Beers Criteria Update Expert Panel. American Geriatrics Society updated Beers Criteria for potentially inappropriate medication use in older adults. J Am Geriatr Soc. 2012; 60:616-31. doi: 10.1111/j.1532-5415.2012.03923.x.

4.      By the American Geriatrics Society Beers Criteria Update Expert Panel. American Geriatrics Society 2019 Updated AGS Beers Criteria(R) for Potentially Inappropriate Medication Use in Older Adults. J Am Geriatr Soc. 2019; 67:674-94. doi: 10.1111/jgs.15767.

5.      By the American Geriatrics Society Beers Criteria Update Expert Panel. American Geriatrics Society 2015 Updated Beers Criteria for Potentially Inappropriate Medication Use in Older Adults. J Am Geriatr Soc. 2015; 63:2227-46. doi: 10.1111/jgs.13702.

6.      Praxedes M, Pereira G, Lima C, Santos DBD, Berhends JS. Prescribing potentially inappropriate medications for the elderly according to Beers Criteria: systematic review. Cien Saude Colet. 2021; 26:3209-19. doi: 10.1590/1413-81232021268.05672020.

7.      Peixoto S, Almeida A, Caramelo A, Mendes L. [Application of the 2015 Beers Criteria Operationalized for Portugal in Institutionalized Elderly: A Cross-Sectional Study]. Acta Med Port. 2021; 34:741-8. doi: 10.20344/amp.13030.

8.      Thiruchelvam K, Byles J, Hasan SS, Egan N, Kairuz T. Frailty and potentially inappropriate medications using the 2019 Beers Criteria: findings from the Australian Longitudinal Study on Women's Health (ALSWH). Aging Clin Exp Res. 2021; 33:2499-509. doi: 10.1007/s40520-020-01772-0.

9.      Abdelwahed AA, El-Dahiyat F, Aljawamis D, Al Ajimi J, Bin Rafeea KJ. Potentially inappropriate medications in older adults according to Beers criteria 2019: Prevalence and risk factors. Int J Clin Pract. 2021; 75:1-8. doi: 10.1111/ijcp.14715.

10.    Tian F, Li H, Chen Z, Xu T. Potentially inappropriate medications in Chinese older outpatients in tertiary hospitals according to Beers criteria: A cross-sectional study. Int J Clin Pract. 2021; 75:1-9. doi: 10.1111/ijcp.14348.

11.    Yang Y, Zhang L, Huang Y, Huang H, Sun S, Xiao J. Based on the Beers Criteria 2019 Edition Over-the-Counter Drugs Risk Confirmation of Elderly Chinese. Biomed Res Int. 2021:1-7. doi: 10.1155/2021/5524551.

12. Drug safety warning Beijing: National Medical Products Administration. [Online] [Cited 2022 October 14]. Available from: URL: https://www.nmpa.gov.cn/yaopin/ypanqjsh/index.html.

13.    Dan W, Jingtian R, Duo D, Lili P, We L. Trend Analysis of Ten-year Annual Reports on Adverse Drug Reaction Monitoring. Chinese J Pharmacovigilance. 2020; 17:276-83. doi: 10.19803/j.1672-8629.2020.05.05.

14.    Cheong VL, Sowter J, Scally A, Hamilton N, Ali A, Silcock J. Medication-related risk factors and its association with repeated hospital admissions in frail elderly: A case control study. Res Social Adm Pharm. 2020; 16:1318-22. doi: 10.1016/j.sapharm.2019.02.001.

15.    Ross MCSB, Wu PE, Atique Md Candidate A, Papillon-Ferland L, Tamblyn R, Lee TC, et al. Adverse Drug Events in Older Adults: Review of Adjudication Methods in Deprescribing Studies. J Am Geriatr Soc. 2020; 68:1594-602. doi: 10.1111/jgs.16382.

16.    Alagiakrishnan K, Wilson P, Sadowski CA, Rolfson D, Ballermann M, Ausford A, et al. Physicians' use of computerized clinical decision supports to improve medication management in the elderly - the Seniors Medication Alert and Review Technology intervention. Clin Interv Aging. 2016; 11:73-81. doi: 10.2147/CIA.S94126.

17.    Prasert V, Shono A, Chanjaruporn F, Ploylearmsang C, Boonnan K, Khampetdee A, et al. Effect of a computerized decision support system on potentially inappropriate medication prescriptions for elderly patients in Thailand. J Eval Clin Pract. 2019; 25:514-20. doi: 10.1111/jep.13065.

18.    Neville HL, Losier M, Pitman J, Gehrig M, Isenor JE, Minard LV, et al. Point Prevalence Survey of Benzodiazepine and Sedative-Hypnotic Drug Use in Hospitalized Adult Patients. Can J Hosp Pharm. 2020; 73:193-201.

19.    Mortensen SJ, Mohamadi A, Wright CL, Chan JJ, Weaver MJ, von Keudell A, et al. Medications as a Risk Factor for Fragility Hip Fractures: A Systematic Review and Meta-analysis. Calcif Tissue Int. 2020; 107:1-9. doi: 10.1007/s00223-020-00688-1.

20.    Fujie K, Kamei R, Araki R, Hashimoto K. Prescription of potentially inappropriate medications in elderly outpatients: a survey using 2015 Japanese Guidelines. Int J Clin Pharm. 2020; 42:579-87. doi: 10.1007/s11096-020-00967-9.

21.    Group SDS, Neurology CSo. Guidelines for diagnosis and treatment of insomnia in Chinese adults. Chin J Neurol. 2012; 45:534-40.

22.    Wang Y, Tan J, Wang L, Pei G, Cheng H, Zhang Q, et al. MiR-125 Family in Cardiovascular and Cerebrovascular Diseases. Front Cell Dev Biol. 2021; 9:1-10. doi: 10.3389/fcell.2021.799049.

23.    Moller H, Aly AF. Research in the field of drug therapy safety management. Br J Clin Pharmacol. 2013; 76:1-4. doi: 10.1111/bcp.12200.

24.    Vega-Morales D, Perez-Luna IRM, Aguirre-Garcia VA, Vazquez- Fuentes BR. Non-Steroidal Anti-Inflammatory Drugs in the Elderly. Agreement With Safe Prescription Recommendations According to Cardiovascular and Gastrointestinal Risks. Reumatol Clin.2021; 17:499-503. doi: 10.1016/j.reuma.2020.05.005.

25.    Joo MK, Park CH, Kim JS, Park JM, Ahn JY, Lee BE, et al. Clinical Guidelines for Drug-Related Peptic Ulcer, 2020 Revised Edition. Gut Liver. 2020; 14:707-26. doi: 10.5009/gnl20246.

26.    Nuzzo A, Brignoli A, Ponziani MC, Zavattaro M, Prodam F, Castello LM, et al. Aging and comorbidities influence the risk of hospitalization and mortality in diabetic patients experiencing severe hypoglycemia. NutrMetab Cardiovasc Dis. 2022; 32:160-6. doi: 10.1016/j.numecd.2021.09.016.

27.    Almomani HY, Pascual CR, Al-Azzam SI, Ahmadi K. Randomised controlled trial of pharmacist-led patient counselling in controlling hypoglycaemic attacks in older adults with type 2 diabetes mellitus (ROSE-ADAM): A study protocol of the SUGAR intervention. Res Social Adm Pharm. 2021; 17:885-93. doi: 10.1016/j.sapharm.2020.07.012.

28.    Zhou JB, Tang X, Han M, Yang J, Simo R. Impact of antidiabetic agents on dementia risk: A Bayesian network meta-analysis. Metabolism. 2020; 109:1-30. doi: 10.1016/j.metabol.2020.154265.

29.    Becker ML, Baypinar F, Pereboom M, Lilih S, van der Hoeven RTM, Giezen TJ, et al. The effect of medication related clinical decision support at the time of physician order entry. Int J Clin Pharm. 2021; 43:137-43. doi: 10.1007/s11096-020-01121-1.

30.    Olakotan OO, Yusof MM. Evaluating the appropriateness of clinical decision support alerts: A case study. J Eval Clin Pract. 2021; 27:868-76. doi: 10.1111/jep.13488.