Table 3 Patients’ self-reported experience of Adverse Drug Events (ADEs)—Multivariable Analysis.

From: A population-based study of self-reported adverse drug events among Lebanese outpatients

Variablea

ORa

Confidence interval

P-value

Geographic area of residence (Beirut is the reference)

Mount Lebanon

0.498

0.262–0.948

0.034

North

2.840

0.502–16.071

0.238

Bekaa

1.105

0.508–2.399

0.802

South/Nabatiyye

0.378

0.091–1.571

0.181

Number of medications taken at home every day

1.225

1.001–1.499

0.049

Intake of any injectable medication

3.008

1.465–6.177

0.003

Asking about possible interactions with prescribed medications while getting over-the-counter medications from the pharmacy

1.945

1.183–3.199

0.009

Physician providing counseling about missing doses

0.439

0.251–0.768

0.004

Physician assessing medication history before prescribing a new medication

1.730

0.947–3.162

0.075

Physician inquiring about previous ADRs before prescribing a new medication

1.599

0.932–2.742

0.088

A member of hospital staff explaining to the patient how to take the medications in an understandable way before hospital discharge

0.430

0.252–0.733

0.002

  1. aVariables with a p-value of 0.2 or less in the bivariate analysis were included in the initial model. Those include: gender; marital status; level of education; geographic area of residence; employment status; number of chronic diseases per patient; chronic kidney disease; intake of any inhaled medication; intake of any injectable medication; initiation of medication-related discussion; not taking a prescribed medication because of counter-indications mentioned in the leaflet and not mentioned by the physician during the visit; asking about possible interactions with prescribed medications while getting the OTC’s from the pharmacy; pharmacist providing counseling about drug interactions, missing doses, accidental overdose, and potential ADRs; physician providing counseling about drug interactions, missing doses, accidental overdose, and potential ADRs; physician assessing medication history before prescribing a new medication; physician inquiring about previous ADRs before prescribing a new medication; hospitalization in the previous year; a member of hospital staff explaining to the patient the purpose of the medications to be taken at home in an understandable way; a member of hospital staff explaining to the patient how to take the medications in an understandable way. Categorical variables identified: level of education, geographic area of residence, employment status, initiation of medication-related discussion.
  2. Using a Backward LR method, the model finally retained the variables shown in this table. Hosmer and Lemshow test for sample adequacy p-value: 0.451. Nagelkerke model summary 0.223.
Back to article page