• 1.

    Pakzad R, Safiri S, 2017. Prevalence of and factors associated with nephropathy in diabetic patients attending an outpatients clinic in Harare, Zimbabwe: methodological issues. Am J Trop Med Hyg 97: 980.

    • Search Google Scholar
    • Export Citation
  • 2.

    Stoltzfus JC, 2011. Logistic regression: a brief primer. Acad Emerg Med 18: 10991104.

Past two years Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 173 62 6
PDF Downloads 36 8 0
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

 

 

Prevalence of and Factors Associated with Nephropathy in Diabetic Patients Attending an Outpatients Clinic in Harare, Zimbabwe: Methodological Issues

Pasipanodya Ian MachinguraDepartment of Medical Laboratory SciencesUniversity of Zimbabwe College of Health SciencesAvondale, Harare, ZimbabweE-mail: imachingura@yahoo.co.uk

Search for other papers by Pasipanodya Ian Machingura in
Current site
Google Scholar
PubMed
Close
,
Exnevia GomoDepartment of Medical Laboratory SciencesUniversity of Zimbabwe College of Health SciencesAvondale, Harare, ZimbabweE-mail: exgomo@gmail.com

Search for other papers by Exnevia Gomo in
Current site
Google Scholar
PubMed
Close
,
Vasco ChikwashaDepartment of Community MedicineUniversity of Zimbabwe College of Health SciencesAvondale, Harare, ZimbabweE-mail: vchikwasha@gmail.com

Search for other papers by Vasco Chikwasha in
Current site
Google Scholar
PubMed
Close
, and
Parmenas Nelson OkwangaWest End ClinicHarare, Zimbabwe E-mail: docpno@gmail.com

Search for other papers by Parmenas Nelson Okwanga in
Current site
Google Scholar
PubMed
Close

Dear Sir:

Thank you for your interest and comments.1 In an earlier draft of our manuscript, we used less stringent variable inclusion criteria in the multivariable model (using a P value < 0.25). During the review and editing process, we decided that the large number of covariates relative to sample size resulted in an unstable outcome, with decreased generalizability beyond the study sample.2 Thus, we included only variables with P < 0.01. We note the limitation of this approach that you highlighted in your letter.

The multivariate analysis data in logistic regression (using a cutoff of P < 0.25) following univariate analysis are given in Tables 1 and 2.

Table 1

Logistic regression analysis of factors associated with overall nephropathy in HIV-negative patients

VariableNormal to mildly increased albuminuria < 30 mg albumin/g creatinineNephropathy ≥ 30 mg albumin/g creatinineUnivariate analysisMultivariate analysis
OR (95% CI)P valueOR (95% CI)P value
Sex
 Female1329710.744N/AN/A
 Male45361.09 (0.65–1.81)
Age, years, mean (SD)56.9 (15.1)59.8 (14.8)1.01 (1.00–1.03)0.0891.01 (0.99–1.03)0.236
Body mass index, kg/m2, mean (SD)27.4 (5.5)25.9 (4.9)0.95 (0.91–0.99)0.0180.97 (0.92–1.02)0.285
Duration of disease, years, mean (SD)9.0 (9.6)13.2 (11.5)1.04 (1.01–1.06)0.0011.02 (0.99–1.05)0.129
Consume alcohol
 Yes1030.39 (0.10-1.43)0.1540.20 (0.04-1.09)0.062
 No16713011
Taking alternative medicine
 Yes13111.14 (0.49–2.63)0.763N/AN/A
 No1641221
HbA1c, %, median (IQR)7.8 (6.4–9.7)8.8 (7.4–11.0)1.21 (1.10–1.33)< 0.0011.13 (0.96–1.32)0.149
Fructosamine, mmol/L3.3 (1.0)3.8 (1.3)1.00 (1.00–1.01)< 0.0011.00 (1.00–1.01)0.165
Triglycerides, mmol/L1.2 (0.8–1.6)1.2 (0.8–1.2)1.11 (0.83–1.47)0.487N/AN/A
Total cholesterol, mmol/L4.6 (3.7–5.4)4.5 (3.6–5.5)1.06 (0.91–1.23)0.456N/AN/A
HDL cholesterol, mmol/L1.1 (0.9–1.4)1.1 (0.9–1.4)1.67 (0.92–3.01)0.0911.25 (0.63–2.49)0.518
Hypertension
 Yes1521211.66 (0.80–3.44)0.1741.17 (0.45–3.01)0.752
 No251211
Retinopathy
 Yes31553.47 (2.06–5.87)< 0.0013.04 (1.70–5.45)< 0.001
 No1417211

CI = confidence interval; HDL = high-density lipoprotein; HIV = human immunodeficiency virus; IQR = interquartile range; N/A = not applicable; OR = odds ratio; SD = standard deviation. In univariate analysis, nephropathy in HIV-negative diabetic patients was significantly associated with lower body mass index (OR 0.95; 95% CI [0.91–0.99]), longer duration of disease (OR 1.04; 95% CI [1.01–1.06]), higher glycosylated hemoglobin (OR 1.21; 95% CI [1.10–1.33]), higher fructosamine (OR 1.00; 95% CI [1.00–1.01]), and retinopathy (OR 3.47; 95% CI [2.06–5.87]). When the variables were subjected to multivariate analysis, only retinopathy (OR 3.04; 95% CI [1.70–5.45]) remained a significant predictor of nephropathy.

Table 2

Logistic regression analysis of factors associated with overall nephropathy in HIV-positive patients

VariableNormal to mildly increased albuminuria < 30 mg albumin/g creatinineNephropathy ≥ 30 mg albumin/g creatinineUnivariate analysisMultivariate analysis
OR (95% CI)P valueOR (95% CI)P value
Sex
 Female91210.483N/AN/A
 Male491.69 (0.39–7.27)
Age, years, mean (SD)52.3 (13.4)53.1 (10.9)1.01 (0.95–1.07)0.835N/AN/A
Body mass index, kg/m2, mean (SD)25.4 (5.7)24.8 (5.0)0.97 (0.85–1.11)0.700N/AN/A
Duration of disease, years, mean (SD)7.2 (8.7)5.5 (6.5)0.97 (0.88–1.06)0.492N/AN/A
Consume alcohol
 Yes01N/AN/AN/AN/A
 No1320
Taking alternative medicine
 Yes20N/AN/AN/AN/A
 No1121
HbA1c, %, median (IQR)7.1 (6.0–9.7)8.5 (6.0–10.4)1.16 (0.87–1.56)0.305N/AN/A
Fructosamine, mmol/L2.7 (0.6)3.8 (1.6)1.01 (1.00–1.02)0.0491.01 (1.00–1.02)0.048
Triglycerides, mmol/L1.1 (0.7–1.3)1.4 (0.9–2.3)2.43 (0.76–7.81)0.1363.07 (0.76–12.43)0.116
Total cholesterol, mmol/L3.8 (3.0–4.7)4.5 (3.7–5.4)1.69 (0.90–3.18)0.1011.43 (0.69–2.99)0.338
HDL cholesterol, mmol/L1.1 (0.8–1.3)1.2 (0.8–1.5)1.35 (0.26–7.13)0.722N/AN/A
Hypertension
 Yes11170.77 (0.12–4.96)0.786N/AN/A
 No241
Retinopathy
 Yes251.67 (0.27–10.33)0.583N/AN/A
 No10151

CI = confidence interval; HDL = high-density lipoprotein; IQR = interquartile range; N/A = not applicable; OR = odds ratio; SD = standard deviation. In univariate analysis nephropathy in HIV-positive diabetic patients was significantly associated with higher fructosamine (OR 1.01; 95% CI [1.00–1.02]). When the variables were subjected to multivariate analysis, only higher fructosamine (OR 1.01; 95% CI [1.00–1.02]) remained a significant predictor of nephropathy. There was an increased odds of nephropathy with increase in fructosamine OR 1.004 (95% CI 1.001–1.007, P = 0.009), controlling for other variables in the model in the manuscript.

REFERENCES

  • 1.

    Pakzad R, Safiri S, 2017. Prevalence of and factors associated with nephropathy in diabetic patients attending an outpatients clinic in Harare, Zimbabwe: methodological issues. Am J Trop Med Hyg 97: 980.

    • Search Google Scholar
    • Export Citation
  • 2.

    Stoltzfus JC, 2011. Logistic regression: a brief primer. Acad Emerg Med 18: 10991104.

Save