The utility of the records medical: factors associated with the medication errors in chronic disease

ABSTRACT Objective: This study describes the development of the medication history of the medical records to measure factors associated with medication errors among chronic diseases patients in Diamantina, Minas Gerais. Methods: retrospective, descriptive observational study of secondary data, through the review of medical records of hypertensive and diabetic patients, from March to October 2016. Results: The patients the mean age of patient was 62.1 ± 14.3 years. The number of basic nursing care (95.5%) prevailed and physician consultations were 82.6%. Polypharmacy was recorded in 54% of sample, and review of the medication lists by a pharmacist revealed that 67.0% drug included at least one risk. The most common risks were: drug-drug interaction (57.8%), renal risk (29.8%), risk of falling (12.9%) and duplicate therapies (11.9%). Factors associated with medications errors history were chronic diseases and polypharmacy, that persisted in multivariate analysis, with adjusted RP chronic diseases, diabetes RP 1.55 (95%IC 1.04-1.94), diabetes/hypertension RP 1.6 (95%CI 1.09-1.23) and polypharmacy RP 1.61 (95%IC 1.41-1.85), respectively. Conclusion: Medication errors are known to compromise patient safety. This has led to the suggestion that medication reconciliation an entry point into the systems health, ongoing care coordination and a person focused approach for people and their families.


Introduction
Chronic disease (CD) is associated with significant morbidity and mortality, and constitutes a substantial burden on the health care system. This is especially true with systemic arterial hypertension and diabetes mellitus, which currently are the most common public health problems, (1) and have a higher burden of disease in Brazil.
Quality patient care is a priority issue in all health care sectors, however, medication errors (ME) are known to compromise patient safety (2) . A ME is any preventable event that may cause or lead to inappropriate medication use or patient harm; this has been studied extensively in developed countries (2)(3) . A systematic review by Tam et al. (4) identified 22 studies, involving a total of 3,755 patients, and found that errors in prescription medication histories occurred in more than 60% of cases. The most important finding of that study was an estimate that 59% of these errors had the potential to cause harm (4) .
Prevention of MEs has therefore become a high priority in patients with CD. Drug-related problems (DRP) may arise at all stages of the medication process, from prescription to treatment follow-up (5) . Therefore, medication reconciliation requires staff to: compile a full list of the patient's previous medications, make a systematic comparison with the active prescriptions, and analyze and resolve any MEs (6) .
Pharmacists are increasingly being recognized as potential partners in many public health activities.
Pharmacists have demonstrated their utility in many areas, including CD management (7) . The involvement of pharmacists in the medication use process, as members of the health care team, improves the quality of patient care by preventing MEs (8)(9) . According to Winter, (8) pharmacists are competent to supervise accurate medication histories and monitoring of error frequencies.
A particularly challenging field is the surveillance of pharmacotherapy in health service, which provides care for CD patients. Polypharmacy has been used in the context of prescribing or taking more medications that are clinically required (10) . Other authors divide the definition of polypharmacy into 'appropriate' and 'problematic' polypharmacy, which the authors of this paper believe supports distinguishing those patients who benefit from multiple medications and those who would benefit from review and reduction of multiple medications (11) . Similarly, health care delivery needs to be structured to improve patient outcomes (12) .
In recent years, the focus of research into optimization of medications for CD patients has shifted from quantitatively measuring the deficiencies in prescribing, to qualitatively uncovering the root causes of suboptimal prescribing (13) . From this research, new avenues for exploration have emerged that may optimize prescribing for CD patients, through targeted interventions and new procedures for medication reviews (13) . One of the most common recommendations after a medication review -discontinuation of medication, or deprescribing -is one of the least likely to be followed (9) . The deprescribing process includes some or all of the following elements: a review of current medications, identification of medications to be discontinued, a discontinuation regimen, involvement of patients, and a review with follow-up (14) . This paper describes the development of the medication history of the medical records to measure factors associated with MEs.

Methods
The subproject was part of a well-defined project Data collection instruments: the material used to evaluate the medication history was a structured form, divided into three parts. The first part included variables on the primary care service and socio-demographic variables, the second part described the primary health care, and the third part described the medications. A pilot test of the form was conducted with ten medical records, as a way to improve the collection instrument.
The medication list identified by the pharmacist was regarded as the most accurate list available in the medical record. The MEs were classified by reviewing the drug that may cause or lead to patient harm. The ME history included: -Polypharmacy: considered as the use of four or more medications (15) .

Results
The medical records of the hypertensive and diabetic patients totaled 396, representing 3.5% of the total family records. The mean age of patients was 62.1 ± 14.3 years, and the minimum and maximum ages of the patients were 25 and 100 years, respectively. Table   1 shows the characteristics of the medical records. In the primary health care, a higher percentage of those over 60 years of age, both men and women, was noted. medications with special characteristics registered is displayed in Figure 1. thioridazine/fluoxetine and metoclopramide/fluoxetine.
In this study, 65.9% of medical records presented drug-to-drug interactions (DDI). A total of 911 potential DDI were identified. Among these, 213 were classified as being potentially major severity, 489 were classified as potentially moderate severity, and 13 were classified as potentially minor severity. There was not one absolutely contraindicated DDI identified in the entire sample.  (21). In the univariate analysis, MEs were associated with these variables: age (p=0.0002), CD (p<0.0001), cerebrovascular disease comorbidity (p=0.0090), and polypharmacy (p<0,0001) but not with sex or number of physician specialties (Table 3).
Factors associated with ME history in the multivariate analysis are presented in Table 4.
Interestingly, association observed between MEs, CD

Discussion
This study demonstrates that a medical record provides the use of medications by diabetic and hypertensive patients, and can be used to assess the impact of primary care management. It can also be used to assess the application of the structured medication history use tool to optimize prescribing, and to reduce MEs. The findings from this study are as follows: (1) Within the medication history obtained by the medical record, a history of MEs was found in the majority of patients (75.7%). Our findings are higher than those of other studies (4,8) .
(2) The majority (35.4%) of the medications are involved with potential DDI. The possibility of DDI (66.2%) detected was higher than that found by other authors (23) (16.3%). This can be attributed to the fact that cardiovascular drugs are the most common drugs to cause DRP (4,24) .
(3) A percentage of medications with renal risk were found (29.8%). This result is probably due to the fact that the study group uses medications with active ingredients that cause nephrotoxicity.
The large difference in percentage of MEs in the medical records as a result is interesting. The Table 3 -(continuation) Cruz, HL, Mota, FKC, Araújo, LU, Bodevan, EC, Seixas, SRS, Santos, DF.
drugs most involved in medication errors, according to the ATC anatomic group were those related to the cardiovascular system, alimentary tract and metabolism and nervous system. A systematic review (4) stated that the prescription drugs most often involved in ME history are cardiovascular agents and sedatives. Another study found mainly antihypertensives were involved (27) .
In absolute numbers, this study found the procedures most recorded were basic nursing care activities (95.5%), and this information is useful to assess the profile of work in primary care, characterized by preventive and curative actions. This profile is in covered areas (29) .
The percentage of medical care present in this study can be used as an indicator of the ability to determine the medication history. This was also discussed by other authors (30) who concluded that the frequency of medication histories taken by physicians is significantly influenced by their specialties. Patients in these specialties are often diagnosed with two or more comorbidities requiring multiple medication therapy (30) . For example, hypertensive patients often have coexisting diabetes, coronary artery disease, or other cardiovascular disorders. This may well explain the percentage of the medication history related to prescription drugs for CD (31) .
Literature on health care service use in Brazil (28)(29)32) has found that the nursing care and physical examination stand out, as in our study. However, there are gaps in the research regarding scientific evidence of medication history, polypharmacy, and DDI in patients with comorbidities.
The majority of DDI in our study were of a moderate severity (53.7%). The most common potential DDI in this group was the interaction between angiotensinconverting enzyme inhibitors and loop diuretics (captorpil/hydrochlortiazide), as noted in another study (33) . Also in this study, 23.4% of the patients were exposed to potentially severe medication combinations.
According to the data, acetylsalicylic acid (25.1%) was the drug that caused such interactions in hypertensive and diabetic patients in another study (24) (28.0%), higher than the 5.3% noted in a different study (33) .
The other aspect of the study was the association between persistent MEs after adjusting for variables, including CDs and polypharmacy. Many studies have described ME rates in hospital settings, but data for primary care is relatively scarce (35 On the other hand, much attention has recently focused on primary care services as the heart of integrated people-centered health care (35) . According to data from one systematic review of 38 studies of primary care interventions, that most successful intervention included a medication review conducted by a pharmacist, leading to a reduction in hospital admissions (39) . Based on data in the literature (35)  There were a few potential limitations to this study.
First, because of the complexity of the medication process and its associated, multifaceted factors, there may have been many other contributing factors to errors that we could not observe or understand. The second of the possible limitations was a study design that was restricted to a quantitative study of the ME history; however a qualitative evaluation of the potential consequences caused by the ME history would have greater clinical relevance. A qualitative investigation, however, was not a pre-defined endpoint of our study.
Potential harm caused by ME history, and consequent medication reconciliation, can only be evaluated in cohort observational study (without reporting of the pharmacist-acquired medication histories) or a randomized trial (pharmacist-versus physician-acquired medication histories). A complete medication history is very time-consuming, and can conceal a medicationrelated problem.

Conclusion
The occurrence of ME in the municipality of Diamantina is a common condition among patients with CD, as is the use of polypharmacy in primary health care. Despite the limitations of the study, it should be highlighted that these factors certainly need to be individually treated in all health care services. In this context, knowing the medication history is important, so that medication reconciliation occurs at the points of entry into the health system.