'Modified STOPP-START criteria for Sri Lanka'; translating to a resource limited healthcare setting by Delphi consensus.

BACKGROUND: 'Screening tool of older people's prescriptions (STOPP) and screening tool to alert to right treatment (START)' criteria is a useful tool to assess the appropriateness of medicines among older adults. However, the original STOPP/START criteria developed in the West, may not be directly applicable to resource limited healthcare settings like Sri Lanka. Hence, we aimed to modify STOPP/START criteria (Version 2) to suit Sri Lanka. METHOD: Two investigators (a clinical pharmacologist and a pharmacist) reviewed and flagged criteria that were unfeasible to Sri Lanka based on their previous research experiences on using STOPP/START version 1. A Delphi consensus methodology was conducted among six experts, including geriatricians, clinical pharmacologists, physicians and a pharmacist, to review and assess each criterion (including the ones flagged by the researchers) for suitability to Sri Lanka. RESULTS: Two Delphi validation rounds were conducted. A final meeting was held with the participation of all experts to resolve disagreements and to establish 100% consensus. The expert panel agreed on a list of 105 criteria, including 70 STOPP and 35 START criteria, indicating an 8% reduction in criteria compared to the original version. Modifications included complete removal (n = 11), re-wording (n = 25), splitting (n = 1) of original criteria and adding a new criterion (n = 1). Main reasons for modifications were unavailability of some medicines in the country, unavailability or inaccessibility of specific clinical information required for assessment of criteria, and adherence to treatment guidelines commonly used in the country. CONCLUSION: A list of 'Modified STOPP/START criteria for Sri Lanka' was developed. These criteria are currently being validated through a multi-centre study.

Usefulness of STOPP/START criteria to assess appropriateness of medicines prescribed to older adults in a resource-limited setting.

Background There is a dearth of published data from resource-limited settings on appropriateness of medicines in older adults using explicit criteria, but it is unclear if the STOPP/START criteria can be helpful. Objective To assess the usefulness of STOPP/START criteria in assessing appropriateness of medicines in a resource-limited setting. Setting Medical, diabetic and psychiatric clinics of a tertiary care hospital and elderly living in a selected locality in Colombo district. Method A descriptive cross-sectional study was conducted over a 2 months period among adults > 60 years on long-term medicines. 'Screening tool of older person's prescriptions' (STOPP) and 'screening tool to alert doctors to right treatment' (START) criteria were used to assess appropriateness of medicines. Main outcome measures Potentially inappropriate medicines (PIMs) and potential prescription omissions (PPOs). Results A total of 468 patients prescribed with 2841 medicines were analysed. PIMs were seen among 167 (35.7%) patients, while PPOs were seen among 289 (61.8%) patients. Incomplete documentation in health records, especially the absence of renal function status (53.7%), was a barrier for accurate assessment of PIMs and PPOs. Some criteria could not be assessed due to differences in the healthcare settings and resources available. Conclusions Inappropriate prescribing to older adults was a problem in the selected settings. Use of explicit criteria to detect inappropriate prescribing is important, but should be modified to suit the local context. Documentation in medical records should be improved to allow for better assessment of appropriateness of medicines.

Pattern of medication errors among inpatients in a resource-limited hospital setting.

PURPOSE OF THE STUDY: There are limited studies on medication errors in South Asian and South East Asian regions. To bridge this gap, we assessed prescribing errors and selected medicine administration errors among inpatients, and the level of acknowledgement of prescribing errors by specialist physicians in a resource-limited hospital setting. STUDY DESIGN: The study was conducted in two medical wards of a hospital in Sri Lanka. Prescribing errors were identified among medicines prescribed in the latest prescription of randomly selected inpatients. Medical notes, medication histories and clinic notes were information sources. Consistency of medicine administration according to prescribing instructions was assessed by matching prescriptions with medicine charts. The level of acknowledgement of prescribing errors by specialist physicians of study wards was assessed by questionnaire. RESULTS: Prescriptions of 400 inpatients (2182 medicines) were analysed. There were 115 patients with at least one medication error. Among the 400 patients, 32.5% (n=130) were prescribing errors. The most frequent types of prescribing errors were 'wrong frequency' (10.3%, n=41), 'prescribing duplications' (10%, n=40), 'prescribing unacceptable medicine combinations' (6%, n=24) and 'medicine omissions' (4.3%, n=17). Medicine charts of 10 patients were inconsistent with prescribing instructions. Wrong medicine administration frequencies were common. The levels of acknowledgment of prescribing errors by the two specialist physicians were 75.5% and 90.9%, respectively. CONCLUSIONS: Prescribing and medicine administration errors happen in resource-limited hospitals. Errors related to dosing regimen and failing to document medicines prescribed or administered to patients in their records were particularly high.

Lipoprotein(a) and lipid profiles of patients awaiting coronary artery bypass graft; a cross sectional study.

BACKGROUND: Lipoprotein(a) (Lp(a)) excess is an independent risk factor of coronary artery disease (CAD) and have shown wide ethnic variations. Further, lipid parameters used in the assessment and management of risk factors for CAD may not reflect accurately the disease or severity if the patients are on pharmacological interventions when compared to Lp(a). Lp(a) levels of Sri Lankan CAD patients awaiting coronary artery bypass graft are not documented. METHODS: A cross sectional study was carried out with patients (n = 102) awaiting coronary artery bypass graft at a tertiary healthcare institution in Sri Lanka. Lp(a) was determined by immunoturbidimetric method (Konelab 20XT) and information on risk factors collected using a standardized questionnaire. The severity of CAD was determined by Gensini score. Lipid parameters and pharmacological treatment data were obtained from the Medical Records. Data were analysed using independent sample t-test, Pearson and Spearman tests respectively. RESULTS: Total cholesterol (TC), LDL cholesterol (LDLc) and HDL cholesterol (HDLc) of the total study sample (average ± SD) were, 150 ± 36 mg/dL, 92 ± 36 mg/dL and 34 ± 9 mg/dL respectively with no significant difference irrespective of being on pharmacological treatment or not. All lipid parameters were significantly high (p < 0.05) in females. The average Lp(a) was 50 ± 38 (SD) mg/dL with no significant difference in males or females independent of being on treatment (50 ± 39 mg/dL) or not (49 ± 39 mg/dL) and above the cut off value (30 mg/dL). CONCLUSIONS: Despite pharmacological interventions 27 % of the study population had high LDLc and majority low HDLc. Mean Lp(a) was in excess irrespective of risk factors or being on treatment or not and is confirmed as an independent, potential marker for assessing the susceptibility for CAD especially in those with other intermediate risk factors but considered non-hyperlipidemic by conventional methods.

Antioxidant potential of green leafy porridges.

OBJECTIVES: To evaluate the antioxidant potential of porridges (kola kenda) made incorporating herbs. DESIGN: Experimental study. Main outcome measures Total phenolic content was estimated by Folin Ciocalteau method (GAEmg/g(FW) and the antioxidant potential by ABTS free radical scavenging activity (TEAC(µg)/100g (FW). RESULTS: Total phenolics and antioxidant potentials of porridges ranged from 5-73 TEAC(µg) /100g and 0.9- 34.2 GAEg/100g respectively. Significantly high (p ≤ 0.05) total phenolic contents were observed in C. auriculata ("Ranawara") [342 GAE mg/g (FW)] and in M. koenigii ("Karapincha") [199 GAEmg/g(FW)]. The highest antioxidant effect was in C. auriculata ("Ranawara") [73 TEAC(µg)/100g(FW)] followed by M. koenigi ("Karapincha") [26 TEAC(µg)/100g (FW)] and C. ternatea ("Ela Katarolu") [24 TEAC(µg)/100g (FW)]. Lowest antioxidant effect was in porridges made with rice and coconut milk both of which had comparable values. A positive correlation (correlation coefficient > 0.8) between total phenolics and the antioxidant potential was observed for all porridges except for Asparagus racemosus (Haathawariya) porridge which had a high antioxidant potential with a low phenolic content. CONCLUSIONS: This study provides evidence that green leafy porridges are a good source of antioxidants in addition to their reported low glycaemic indices.

The pattern of abbreviation use in prescriptions: a way forward in eliminating error-prone abbreviations and standardisation of prescriptions.

INTRODUCTION AND OBJECTIVES: Inappropriate abbreviations used in prescriptions have led to medication errors. We investigated the use of error-prone and other unapproved abbreviations in prescriptions, and assessed the attitudes of pharmacists on this issue. METHODS: A reference list of error-prone abbreviations was developed. Prescriptions of outpatients and specialty clinic patients in a teaching hospital in Sri Lanka were reviewed during one month. An interviewer administered questionnaire was used to assess attitudes of pharmacists. RESULTS: 3370 drug items (989 prescriptions) were reviewed. The mean (standard deviation) number of abbreviations per prescription was 5.9 (3.5). The error-prone abbreviations used in the hospital were, µg (microgram), mcg (microgram), u (units), cc (cubic centimeter), OD (once a day), @ sign, d (days/daily), m (morning) and n (night), and among all prescriptions reviewed, they were used at a rate of 17.4%, 0.1%, 1.9%, 0.2%, 0.2%, 4.9%, 23.5%, 4.4% and 15.8% respectively. Among the 103 types of abbreviations observed, 71 were not standard acceptable abbreviations. Multiple abbreviations were used to indicate a single drug item/ instruction (N = 7). The abbreviation 'd' was used to denote 'daily' as well as 'days'. All pharmacists believed that using error-prone abbreviations will always (5.3%) or sometimes (94.7%) lead to medication errors. CONCLUSIONS: Error-prone abbreviations and many other unapproved abbreviations are frequently used in hospitals. There is a need to educating health care professionals on this issue and introduce an in-house error-prone abbreviation list for their guidance.