Goals

The objective of this paper is to exploit the database of HIV hospital admissions in Spain during the period 1997-2015. The main goals are:

Available data

We conducted an observational retrospective study using the SNHDD [11],[12]. The SNHDD includes information on the sex, age, dates of admission and discharge, up to 14 discharge diagnoses, and up to 20 procedures performed during the hospitalization. Data were collected from January 1, 1997, to December 31, 2015 (19 years). Country of origin is not available in the SNHDD. The SMH conducts regular audits to assess the accuracy of the SNHDD [11],[13]. The database can be accessed upon request and data are only anonymously given.

The International Classification of Diseases-Ninth Revision, Clinical Modification (ICD-9-CM) is used in the SNHDD. which is the one used by the Spanish SNHDD. We selected hospital admissions for patients with a diagnosis within the ICD-9-CM code 042 [Human immunodeficiency virus [HIV] disease] and V08 [Asymptomatic human immunodeficiency virus [HIV] infection status] from 1997 to 2015 at any position in the diagnostic list for each episode of hospital admission.

Regardless the position within the diagnoses coding list, we retrieved data about comorbid specific conditions using the enhanced ICD-9-CM. The list of ICD-9CM and comorbidities is in Table 1.

We further examined several hospital outcome variables, including re-admissions (if the patient had been discharged within the previous 30 days after prior registration in the same clinic), and in-hospital mortality according to the SNHDD of the Spanish Ministry of Health

The available database consists on 472269 entries from 1997 until 2015. The volume by year, gender and region is shown in Figures 1 and 2. The territory in Spain is divided in a total of 19 political regions, explicit in Figure 2.

Statistical analysis.

The available data consists on dessagregated data from VIH individuals being admitted to hospitals during the period 1997-2015. The database does not contain any identifier for each person, so it is impossible to assume independency between data: the same person probably will have several entries to the hospital during a year and also along the years during the period of interest.

Given the high variability in available data, coming from individuals and also political regions in Spain, in order to study trends along the years, data will be aggregated by years, in terms of counts (for categorical variables) or means (for continuous variables). For visualization, plots of means for continuous variables and ratios for count data are used.

We’ll focus on studying differences due to gender, and sometimes region (acting as main effects), as well as its evolution along the years under the period of interest. Evolution of the observed trend for different groups (Males/Females) along the years will be studied throug interaction effects. When the response variable is continuous, linear regression models will be fitted. When the response variable consists on counts or proportions, generalized linear models will be fitted, assuming a binomial likelihood (and link logit: \(log(p/(1-p))\)) is the response is dicotomous, or poisson (and link \(log\)) when it is not associated to a dicotomous variable. The independency assumptions of linear and generalized linear models will lack irreversibly, as commented before, but will be undertaken.

Variable Year will be treated as covariable in order to provide the effect of change in time along the period of interest. It will be differenced to start in t=1 for 1997 and finish at t=15 por year 2015, in order to provide a clear effect of a unit change in time on the response (it will be referred to as NYear.from.1996=number of years from 1996). When the observed trend time-response is non-linear, the year variable will entry the model through a cubic spline transformation. A cubic spline model is a nonparametric regression technique defined piecewise by polynomials, and used in the literature to address longitudinal data fitting problems, like those we are concerned in this paper. It has been profusely used in applied statistic studies, even in Medicine (see Durrleman and Simon, 1989, or Desquilbet and Mariotti, 2010, among many others). Splines models are adaptive in the sense that fit the data in a continuous way, by dividing data in segments and assessing the fit by taking into acount the data from the closest entries (in our case from the closest years). Knots defining segments are defined in terms of the quantiles of the data. The number of knots, or equivalently the degrees of freedom, must be chosen dependent on the data. A fit where the predictor x enters through a spline model with df=1 is equivalent to the usual linear model; in this case, for simplicity of interpretations, a usual linear model will be fitted. A model with df=2 implies dividing the x values in 2 segments (through the median as a knot) and in some way similar to fit a polynomial of degree 2 at each segment; a model with df=3 implies dividing the x values in 3 segments and fitting something similar to a polynomial of degree 3 at each segment. See details in De Boor (1978) and Welham et al (2007). The natural splines models are implemented with the library splines in R (ns).

Model selection -selection of significative effects and also the number of knots for the splines- will be resolved in terms of step searchs based on the AIC criteria. The AIC, Akaike Information Criterion, (Akaike, 1974), is used as a model selection index that takes into account the likelihood of the proposed model for the fitted model, and also the number of parameters used for the fit, so promoting simpler models that provide enough likelihood. Goodness of fit is reported in terms or R-squared indexes for linear models and significance of the full model based on the F-statistic for the residuals. For generalized linear models, the percentage of deviance explained by the model, 1-Residual.Deviance/Null.Deviance, could reproduce the interpretation of R-squared in the linear model, and it will be reported. Significance of effects will also be reported in this paper in order to justify differences by the variables of interest.

The R statistical package version 3.6.3 (The R Foundation) was used to perform the statistical analysis.

Evolution of the rates of VIH admissions by gender

In this study 66.451.094 hospital admissions were recorded in Spain during 19 years (1997-2015), 52% of them concerning females. In the VIH database, an overall of 472.269 admissions of VIH patients were registered for that period, with just the 27.5% of females. Rates of admissions were assesed, both for males and females, by dividing the number of entries in the VIH database, by the total number of hospital admissions in Spain. A model

Figure 3 shows the relative counts of VIH admissions per 10.000 hospital admissions in Spain, distinguised for males and females. In order to statistically prove the impact of time on these volumes, a binomial logit generalized linear model has been fitted to explain the differences by gender, and minimum complexity cubic splines have been considered to include the Year effect. The goodness of fit is checked and confirmed with an AIC of 1013,9 and a 99.5% percentage of total deviance explained by the model. In the fitted model, variable “Year” results significative and interacting with “Sex”, to explain the decrease in the relative volume of VIH admissions along the time (the p-value for the interaction is \(1.6 \cdot 10^^ {-12}\)). The fitted model results in a negative prediction line to explain the evolution of rate admissions along the years. As it can be seen in Figure 3, the slope of the fitted model (lines) is smaller (in absolute value) for females than males: female rates decrease less than males along the years. Intercept is also smaller for females: the VIH group has a smaller weight on hospital admissions within females than within males. \[\mbox{logit(proportion for males) } = -4.24-0.0268 \times NYear.from.1996 \]

\[\mbox{logit(proportion for females) } = -5.27 -0.0225 \times NYear.from.1996.\]

Evolution of VIH admissions by Region

Although differences due to gender have been proven on the rates of VIH on general hospital admissions, now we want to investigate gender differences in every region in Spain, concerning the rate of VIH hospital admissions per 100.000 inhabitants of the same sex. We have acceeded the data on the population of the different regions in Spain during the period of interest, both for males and females (source INE). Then the ratios of VIH admissions per 100.000 inhabitants of the same gender have been assessed and represented in Figure 4 (dots).

We can observe from Figure 4 that population of VIH admissions on male population has a greater impact than on female population, whatever the region in Spain: male curves are above female’s. In order to fit the time trend, both for females and males, in all regions, we use a binomial GLM model with response the rates and enter the Year variable through a spline of degree 3 (superior in goodness of fit to degrees 1 or 2). The AIC of the fitted model results in 7570.767 and the percentage of deviance explained is 98.82%. All effects and interactions are significant (with p-values \(<2.2 \cdot 10{-16}\)). So, time effect, as well as gender effect is significant in providing differences on the impact of VIH in the general population, differently for males and females. The rate of VIH admissions does not evolve linearly along the years, but in most regions is quite estable, as shown in Figure 4. Remarkable decreases along time (comparing 1997 and 2015) of male rates produce in Andalucía, Baleares, Cantabria, Cataluña, Ceuta, Comunidad Valenciana, Madrid, Melilla and Pais Vasco.

Evolution of mean age differenciated by sex

As shown in Figure 5, mean age of VIH patients registered in the hospital admissions database (represented by dots), have uniformly grown up, both for males and females, during the period of interest. Vertical lines represent the interquartile (IQ) range (between cuantiles q0.25 amd q0.75) for males and females every year; a large variability can be appreciated.

A linear regression model has been fitted to the means, by entering the year (initiated in year=1) as a covariate.

## $$
## \text{m} = 34.04 + 0.81(\text{anyo}) - 3.94(\text{Sexo}_{\text{Female}}) + 0.03(\text{anyo} \times \text{Sexo}_{\text{Female}}) + \epsilon
## $$

The index \(R^2= 0.9984\) and the goodness of fit F-test gives a p-value \(<2.2 \cdot 10{-16}\), so giving confidence on a good fitting. Slope of the prediction line (joining the means in Figure 5) for females is slightly higher than for males though significatively different (p-value \(< 8.4 \cdot 10^{-9}\) ). The time effect (one year) on the response (the mean age) is 0.811 for males and 0.845 form females.

Evolution of Type of admission for VIH patients

In Figure 6 the counts of emergency, scheduled and other admissions are shown for VIH hospital admissions, differentiated for males and females. Though scheduled admissions have not varied along the period of interest, emergency admissions show a slight decrease, specially for males before and after year 2000. We test for differences in the proportions for these two periods through a Chi-squared test, both for males and females. This test results in a significant p-value \(<2.2 \cdot 10{-16}\) for males, but insignificant for females (p-value \(=0.145\)), so we can conclude that these two periods have produced different emergency admissions rates just on the group of males.

Evolution of Exitus rate for VIH admissions.

Though EXITUS is not very relevant as a cause of discharging from the hospital, as shown in Figure 7a, it is studied in order to check for differences between males and females and study its change along the period of interest. In Figure 7b, Exitus rates for males and females are shown.

Prevalence of diagnoses

Diagnoses of AIDS-defining diseases

Next we study prevalence and evolution of diagnosis AIDS-defining diseases (see Table with codes), as well as possible differences in incidence for males and females.

From the available data, there is no any case of “Coccidioidomycosis, disseminated or extrapulmonary”, so this disease is excluded from the analysis below.

## Note: Using an external vector in selections is ambiguous.
## ℹ Use `all_of(indices)` instead of `indices` to silence this message.
## ℹ See <https://tidyselect.r-lib.org/reference/faq-external-vector.html>.
## This message is displayed once per session.
Table.p1. Prevalence of diagnoses in % per year. AIDS-defining diseases.
disease 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
tuberc 15.4 13.9 12.8 12.6 11.7 12.1 11.8 11.9 12.7 12.3 12.8 13.5 13.7 13.1 13.8 14.8 14.7 14.7 14.7
pneumocys 6.2 5.5 4.3 4.1 3.6 3.9 3.5 3.2 3.1 3.1 2.9 2.7 2.9 2.8 2.8 3.0 2.7 2.6 2.3
toxoplas 4.2 3.2 2.8 2.4 2.3 2.0 1.9 1.8 1.8 1.8 1.6 1.5 1.5 1.2 1.2 1.3 1.1 1.1 0.8
cytomega 3.0 1.9 1.4 1.4 1.3 1.1 0.9 1.1 1.0 1.0 0.9 0.9 1.0 1.0 1.2 1.4 1.2 1.1 1.2
candieso 2.3 2.5 2.3 2.6 2.5 2.3 2.5 2.5 2.5 2.4 2.3 2.2 1.9 1.7 1.7 1.6 1.6 1.5 1.5
kaposi 1.9 1.5 1.3 1.2 1.1 1.3 1.1 1.1 1.0 1.1 1.2 1.0 1.1 1.3 1.1 1.3 1.3 1.2 1.2
progleukoen 1.9 2.0 1.7 1.6 1.8 1.5 1.5 1.7 1.5 1.6 1.6 1.4 1.3 1.2 1.2 1.2 1.2 1.1 1.3
wasting 1.7 1.7 1.9 1.9 1.8 2.0 2.3 2.3 2.0 2.1 1.9 2.4 2.1 2.1 1.9 1.8 1.7 1.7 1.7
mycobac 1.5 1.0 1.0 1.1 0.9 0.9 0.8 0.7 0.8 0.7 0.6 0.5 0.6 0.5 0.5 0.4 0.5 0.5 0.4
herpes 1.2 1.0 1.0 0.9 0.8 0.9 0.9 0.8 0.9 0.9 0.8 0.8 0.7 0.6 0.7 0.6 0.6 0.6 0.6
enceph 1.1 0.8 0.8 0.7 0.6 0.7 0.8 0.7 0.8 1.0 1.0 0.8 0.9 1.1 1.0 1.0 0.9 1.1 1.0
crytococ 0.9 0.7 0.6 0.6 0.6 0.6 0.5 0.5 0.5 0.6 0.5 0.4 0.3 0.4 0.4 0.3 0.3 0.3 0.3
lymburkit 0.5 0.5 0.5 0.5 0.5 0.4 0.5 0.6 0.7 0.9 0.6 0.6 0.8 1.1 0.9 1.2 1.0 0.9 0.9
isospor 0.2 0.3 0.1 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
leishman 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.3 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.1
salmon 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0
candiresp 0.1 0.1 0.1 0.2 0.1 0.1 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.1
histoplas 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.1 0.1 0.0 0.0 0.0 0.0 0.0
cervican 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0
crytospor 0.0 0.0 0.3 0.3 0.3 0.2 0.2 0.2 0.2 0.4 0.3 0.2 0.1 0.1 0.1 0.1 0.1 0.1 0.1
lymbrain 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.1 0.0 0.0 0.0 0.1

EN FIG.P2 ESTÁN PINTADOS LOS DATOS, en FIG.PP2 ESTÁN PINTADOS LOS AJUSTES DE SUAVIZADO Y LOS PUNTOS.

In Fig.p2 all diseases are represented with their prevalence. For those diseases presenting low prevalence, below 2%, their behavior along the period 1997-2015 is not clear from the figure. In Fig.p3 just the diseases with prevalence above 2% some year of the period, are represented. The common trend in all of them, excluding Tuberculosis, has been to decrease in time.

Table 2 shows the prevalence of each diagnosis, assessed as the percentage of diagnoses produced every year. Tuberculosis is the more prevalent diagnosis, and also one of the diagnoses that has experimented a mayor change along the period of study: with prevalences around 15% in 1997, decreased appearance till 2003 and from there it has been increasing to prevalence again rounding 15%, which has maintained constant during 2012-2015.

In Fig.pp2 the behavior along time for every definitory illness of AIDS IS shown. Trends have been fitted through binomial GLMs for proportion of cases in a year, including the effect of time through a spline of degree 3 interacting with disease. The model gives statistical evidence of the time effect on the prevalence of every disease, with a deviance explained with respect to the null model of 99.708% and an AIC of 3552.628. Just for Progressive multifocal leukoencephalopathy the trend along time is linear, decreasing with time.

Behavior for the most prevalent Diseases defining AIDS

EN FIG.P3 ESTÁN PINTADOS LOS DATOS, en FIG.PP3 ESTÁN PINTADOS LOS AJUSTES DE SUAVIZADO Y LOS PUNTOS.

In Fig.pp3 the behavior along time for the most prevalent definitory illness of AIDS is shown. Trends have been fitted through binomial GLMs for proportion of cases in a year, including the effect of time through a spline of degree 3 interacting with disease. The model gives statistical evidence of the time effect on the prevalence of every disease, with a deviance explained with respect to the null model of 99.733% and an AIC of 1621.879. Just for Progressive multifocal leukoencephalopathy the trend along time is linear, decreasing with time.

Non AIDS-defining diseases

Table p4. Prevalence of diagnoses in % per year.
disease 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Parental Drug Addiction 36.3 40.8 43.0 43.9 45.6 44.8 50.9 51.1 52.2 51.2 51.7 53.0 52.6 52.7 51.9 51.4 51.3 51.6 50.0
Pneumonia (recurrent) 13.1 13.3 14.3 13.4 12.6 12.6 13.0 12.2 13.8 12.6 13.8 13.2 12.4 10.9 10.3 9.6 9.3 9.5 9.0
Hepathitis C Virus 11.3 15.2 26.3 32.0 36.4 39.2 45.8 47.6 48.8 49.3 48.7 49.2 48.1 47.9 47.8 46.9 46.9 47.1 44.9
Hepathitis B Virus 3.2 3.8 7.6 8.8 9.1 9.0 10.3 10.5 9.7 9.9 9.0 9.0 9.3 8.8 8.8 8.0 8.0 7.6 7.5

Prevalence by five-year-term

Estas tablas a continuación no sería necesaria dados los resultados mostrados arriba. Como la había hecho, te la dejo, pero yo no la incluiría.

Table pp3. Diagnosis prevalence per 100 admissions and Chi-squared test. AIDS-defining diseases.
[1997,2001] (2001,2006] (2006,2011] (2011,2015] Chi2 p-value
Candidiasis esophageal 2.46 2.42 1.97 1.56 0.000
Candidiasis of bronchi, trachea or lungs 0.14 0.19 0.20 0.18 0.003
Cervical cancer (invasive) 0.02 0.02 0.03 0.03 0.013
Crytococcosis, extrapulmonary 0.68 0.52 0.41 0.30 0.000
Cryptosporididiosis, chronic intestinal for longer than 1 month 0.18 0.26 0.16 0.13 0.000
Cytomegalovirus disease (other than liver, spleen or lymph nodes) included retinitis 1.79 1.03 1.00 1.24 0.000
Encephalopathy (HIV-related) 0.79 0.81 0.96 0.97 0.000
Herpes simplex: chronic ulcer(s) (for more than 1 month); or bronchitis, pneumonitis or esophagitis 0.95 0.86 0.74 0.60 0.000
Histoplasmosis, deisseminated or extrapulmonary 0.02 0.04 0.05 0.04 0.000
Isosporiasis, chronic intestinal (for more than 1 month) 0.14 0.02 0.04 0.04 0.000
Kaposi’s sarcoma 1.41 1.10 1.12 1.26 0.000
Leishmaniasis visceral 0.22 0.21 0.21 0.17 0.049
Lymphoma primary brain 0.00 0.00 0.03 0.05 0.000
Lymphoma Burkitt’s, or immunoblastic 0.49 0.62 0.79 0.99 0.000
Mycobacterium avium complex and Mycobacterium, other species, disseminated or extrapulmonary 1.11 0.78 0.55 0.45 0.000
Pneumocystitis carinii pneumonia 4.73 3.36 2.80 2.65 0.000
Progressive multifocal leukoencephalopathy 1.80 1.56 1.36 1.19 0.000
Salmonella sepsis (recurrent) 0.13 0.11 0.05 0.02 0.000
Toxoplasmosis of the brain 2.98 1.85 1.41 1.06 0.000
Tuberculosis 13.29 12.16 13.37 14.73 0.000
Wasting syndrome due to HIV 1.80 2.12 2.07 1.71 0.000
Table pp4. Diagnosis prevalence per 100 admissions and Chi-squared test.
[1997,2001] (2001,2006] (2006,2011] (2011,2015] Chi2 p-value
Hepathitis B Virus 6.53 9.87 8.98 7.78 0
Hepathitis C Virus 24.32 46.15 48.38 46.47 0
Pneumonia (recurrent) 13.32 12.85 12.22 9.33 0
Parental Drug Addiction 41.92 50.06 52.39 51.11 0

Sex differences

When we try to study the effect of sex on these prevalences, again we encounter differences clearly visible in Fig.p5.

In order to statistically verify these differences, again we take the more prevalent diseases and fit the trending model fitted above, but including the effect of year, also interacting with time and disease. Significative interactions will denote different behavior between genders, for different diseases along time.

In Fig.pp5 the behavior along time for every definitory illness of AIDS IS shown, together with the observed differences by sex. Trends have been fitted through binomial GLMs for proportion of cases in a year within the sex group, including the effect of time through a spline of degree 3 interacting with disease and year. The model proves significative, with a deviance explained with respect to the null model of 99.616% and an AIC of 2114.846. Just for Progressive multifocal leukoencephalopathy the trend is linear, both for males and females, decreasing with time and with a marked effect of sex shown in the slope of the regression lines.

Association in diagnoses

In order to investigate which diseases are more correlated as diagnostiscs in hospital admissions for the VIH patients, we assess the number of cases that two diseases are diagnosed jointly for any patient admitted during the period of interest. Then, Figures Fig.p7 (Fig.p8) show the absolute (relative) frequencies, for all diseases defining AIDS, and also for Hepatitis B and C.

HABRÍA QUE ELEGIR ENTRE VISUALIZAR FIG.P7 O FIG.Pp7. YO OPTARÍA POR FIG.pp7 CON LAS FRECUENCIAS RELATIVAS. LOS DATOS ESTÁN EN LA TABLE.P7, desglosados en dos tablas.

Table.p7a. Cases per 10.000 where both diseases are jointly diagnosed at admission. AIDS-defining diseases (part I).
candiresp cervican crytococ crytospor cytomega enceph herpes histoplas isospor kaposi
candieso 0.72 0.02 1.14 2.33 10.29 3.07 5.55 0.19 0.59 4.79
candiresp NA 0.00 0.08 0.02 0.78 0.23 0.40 0.00 0.02 0.30
cervican NA NA 0.00 0.00 0.00 0.00 0.04 0.00 0.00 0.00
crytococ NA NA NA 0.25 1.80 0.61 0.70 0.02 0.13 1.76
crytospor NA NA NA NA 1.40 0.34 0.40 0.02 0.06 0.87
cytomega NA NA NA NA NA 1.86 2.71 0.19 0.36 5.93
enceph NA NA NA NA NA NA 0.72 0.08 0.13 1.06
herpes NA NA NA NA NA NA NA 0.25 0.17 1.86
histoplas NA NA NA NA NA NA NA NA 0.00 0.11
isospor NA NA NA NA NA NA NA NA NA 0.28
kaposi NA NA NA NA NA NA NA NA NA NA
leishman NA NA NA NA NA NA NA NA NA NA
lymbrain NA NA NA NA NA NA NA NA NA NA
lymburkit NA NA NA NA NA NA NA NA NA NA
mycobac NA NA NA NA NA NA NA NA NA NA
pneumocys NA NA NA NA NA NA NA NA NA NA
progleukoen NA NA NA NA NA NA NA NA NA NA
salmon NA NA NA NA NA NA NA NA NA NA
toxoplas NA NA NA NA NA NA NA NA NA NA
tuberc NA NA NA NA NA NA NA NA NA NA
Table.p7b. Cases per 10.000 where both diseases are jointly diagnosed at admission. AIDS-defining diseases (part II).
kaposi leishman lymbrain lymburkit mycobac pneumocys progleukoen salmon toxoplas tuberc wasting
candieso 4.79 0.74 0.04 0.74 6.16 25.62 3.49 0.57 9.72 35.51 19.10
candiresp 0.30 0.02 0.02 0.08 0.19 3.07 0.30 0.08 0.25 2.63 1.06
cervican 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.23 0.00
crytococ 1.76 0.17 0.02 0.15 0.93 3.07 0.83 0.19 1.72 6.12 1.59
crytospor 0.87 0.11 0.02 0.02 0.72 1.46 0.15 0.00 0.55 2.69 2.44
cytomega 5.93 0.34 0.15 1.55 5.61 20.67 2.05 0.19 8.49 16.03 6.12
enceph 1.06 0.08 0.02 0.17 1.23 2.69 2.39 0.06 2.12 9.68 4.57
herpes 1.86 0.19 0.04 1.19 1.16 11.62 0.83 0.15 3.45 10.21 3.07
histoplas 0.11 0.06 0.00 0.06 0.15 0.17 0.00 0.00 0.02 0.23 0.15
isospor 0.28 0.02 0.00 0.02 0.19 0.47 0.04 0.00 0.36 0.80 0.47
kaposi NA 0.64 0.06 1.40 2.56 9.00 2.14 0.13 4.57 15.67 2.94
leishman NA NA 0.00 0.32 0.42 0.61 0.19 0.04 0.85 2.60 0.66
lymbrain NA NA NA 0.00 0.04 0.04 0.08 0.00 0.13 0.15 0.15
lymburkit NA NA NA NA 0.25 0.57 0.13 0.02 0.38 5.10 0.70
mycobac NA NA NA NA NA 3.35 1.27 0.08 2.82 8.32 6.56
pneumocys NA NA NA NA NA NA 3.66 0.85 6.71 38.79 14.63
progleukoen NA NA NA NA NA NA NA 0.19 5.99 14.38 4.21
salmon NA NA NA NA NA NA NA NA 0.19 0.91 0.42
toxoplas NA NA NA NA NA NA NA NA NA 27.53 5.93
tuberc NA NA NA NA NA NA NA NA NA NA 33.46
Table.p8. Cases per 10.000 where both diseases are jointly diagnosed at admission (part I).
hepac pneumo pardrug candieso candiresp cervican crytococ crytospor cytomega enceph herpes histoplas
hepab 576.11 102.23 527.98 14.21 1.55 0.23 2.71 0.89 5.76 5.95 5.53 0.17
hepac NA 521.74 2631.76 76.65 7.14 1.19 10.63 4.30 21.70 34.15 26.09 0.19
pneumo NA NA 717.81 33.62 5.84 0.06 3.85 1.74 21.53 11.48 22.21 0.08
pardrug NA NA NA 105.22 9.53 0.80 16.13 6.52 36.65 43.64 38.66 0.44
candieso NA NA NA NA 0.72 0.02 1.14 2.33 10.29 3.07 5.55 0.19
candiresp NA NA NA NA NA 0.00 0.08 0.02 0.78 0.23 0.40 0.00
cervican NA NA NA NA NA NA 0.00 0.00 0.00 0.00 0.04 0.00
crytococ NA NA NA NA NA NA NA 0.25 1.80 0.61 0.70 0.02
crytospor NA NA NA NA NA NA NA NA 1.40 0.34 0.40 0.02
cytomega NA NA NA NA NA NA NA NA NA 1.86 2.71 0.19
enceph NA NA NA NA NA NA NA NA NA NA 0.72 0.08
herpes NA NA NA NA NA NA NA NA NA NA NA 0.25
histoplas NA NA NA NA NA NA NA NA NA NA NA NA
isospor NA NA NA NA NA NA NA NA NA NA NA NA
kaposi NA NA NA NA NA NA NA NA NA NA NA NA
leishman NA NA NA NA NA NA NA NA NA NA NA NA
lymbrain NA NA NA NA NA NA NA NA NA NA NA NA
lymburkit NA NA NA NA NA NA NA NA NA NA NA NA
mycobac NA NA NA NA NA NA NA NA NA NA NA NA
pneumocys NA NA NA NA NA NA NA NA NA NA NA NA
progleukoen NA NA NA NA NA NA NA NA NA NA NA NA
salmon NA NA NA NA NA NA NA NA NA NA NA NA
toxoplas NA NA NA NA NA NA NA NA NA NA NA NA
tuberc NA NA NA NA NA NA NA NA NA NA NA NA
Table 8pp. Cases per 10.000 where both diseases are jointly diagnosed at admission (part II).
histoplas isospor kaposi leishman lymbrain lymburkit mycobac pneumocys progleukoen salmon toxoplas tuberc wasting
hepab 0.17 0.32 8.19 1.84 0.08 6.06 4.51 19.12 10.44 0.72 10.40 104.75 14.72
hepac 0.19 0.80 17.24 7.88 0.64 12.03 21.09 82.22 55.50 2.77 44.45 483.01 79.62
pneumo 0.08 0.57 11.29 1.86 0.08 3.73 6.56 24.92 11.71 1.14 13.55 110.11 36.06
pardrug 0.44 1.55 30.79 8.41 0.64 18.38 29.81 152.50 67.84 3.79 70.09 679.91 109.58
candieso 0.19 0.59 4.79 0.74 0.04 0.74 6.16 25.62 3.49 0.57 9.72 35.51 19.10
candiresp 0.00 0.02 0.30 0.02 0.02 0.08 0.19 3.07 0.30 0.08 0.25 2.63 1.06
cervican 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.23 0.00
crytococ 0.02 0.13 1.76 0.17 0.02 0.15 0.93 3.07 0.83 0.19 1.72 6.12 1.59
crytospor 0.02 0.06 0.87 0.11 0.02 0.02 0.72 1.46 0.15 0.00 0.55 2.69 2.44
cytomega 0.19 0.36 5.93 0.34 0.15 1.55 5.61 20.67 2.05 0.19 8.49 16.03 6.12
enceph 0.08 0.13 1.06 0.08 0.02 0.17 1.23 2.69 2.39 0.06 2.12 9.68 4.57
herpes 0.25 0.17 1.86 0.19 0.04 1.19 1.16 11.62 0.83 0.15 3.45 10.21 3.07
histoplas NA 0.00 0.11 0.06 0.00 0.06 0.15 0.17 0.00 0.00 0.02 0.23 0.15
isospor NA NA 0.28 0.02 0.00 0.02 0.19 0.47 0.04 0.00 0.36 0.80 0.47
kaposi NA NA NA 0.64 0.06 1.40 2.56 9.00 2.14 0.13 4.57 15.67 2.94
leishman NA NA NA NA 0.00 0.32 0.42 0.61 0.19 0.04 0.85 2.60 0.66
lymbrain NA NA NA NA NA 0.00 0.04 0.04 0.08 0.00 0.13 0.15 0.15
lymburkit NA NA NA NA NA NA 0.25 0.57 0.13 0.02 0.38 5.10 0.70
mycobac NA NA NA NA NA NA NA 3.35 1.27 0.08 2.82 8.32 6.56
pneumocys NA NA NA NA NA NA NA NA 3.66 0.85 6.71 38.79 14.63
progleukoen NA NA NA NA NA NA NA NA NA 0.19 5.99 14.38 4.21
salmon NA NA NA NA NA NA NA NA NA NA 0.19 0.91 0.42
toxoplas NA NA NA NA NA NA NA NA NA NA NA 27.53 5.93
tuberc NA NA NA NA NA NA NA NA NA NA NA NA 33.46
  • Hepatitis B is very correlated to Hepatitis C, pneumo, pardrug, tuberc
  • Hepatitis C is very correlated to pneumo, pardrug (very remarkable), tuberc, wasting, pneumocys, progleukoen.
  • pneumo is very associated to pardrug, tuberc y wasting
  • pardrug is very associated to tuberc, wasting, pneumocys, toxoplas, progleukoen, enceph, herpes, cytomega
  • candieso is very associated to tuberc

DEPENDE DE DÓNDE MARQUEMOS UN CRITERIO DE ASOCIACIÓN: EJ: SUPERIOR A 30/10.000 o 100/10.000

In Fig 8p/8pp and Table 8p/8pp the prevalence -in number of cases per 10.000- where both diseases are jointly diagnosed at hospital admission are displayed.

SI ESTAMOS INTERESADOS EN RATIFICAR ESTAS ASOCIACIONES, HABRÍA QUE HACER ODDS-RATIO (A FAVOR DE QUE APAREZCAN CONJUNTAMENTE). CONFÍRMAME SI QUIERES QUE PROCEDA, PUES HE DE HACERLO PARA CADA PAREJA.

Prevalence of AIDS-defining diseases between 1997 and 2015

We have assessed a new variable diaids that takes the value “1” whenever an admission has been diagnosed with any of the AIDS-defining diseases, and “0” otherwise.

The prevalence of any AIDS-defining diseases is displayed in Fig8, both assessed for males and females separately. Again this prevalence is lesser for females than for males, though the trend is similar, quickly decreasing in a cuadratic way from 1997 to 2004, and becoming stabilized from 2004 on.

A spline model with 4 degrees of freedom is fitted, in order to suit this nonlinear behavior along time. Interaction with Sex variable has been included in the model, and has resulted significant. The explained deviance by the model is 95.74% and the AIC is 485.6.

This model justifies the effect of time on AIDS-defining diseases, and the differences of prevalence on the male group, respect to the female one.

References

De Boor, C. (1978). A practical guide to splines. New York : Springer. https://www.springer.com/gp/book/9780387953663

Welham, S.J., Cullis, B.R., Kenward, M.G., Thompson, R. (2007). A comparison of mixed model splines for curve fitting. Australian and New Zealand Journal of Statistics, 49,1, 1-23. DOI https://doi.org/10.1111/j.1467-842X.2006.00454.x

Desquilbet, L. and Mariotti, F. Dose‐response analyses using restricted cubic spline functions in public health research, Statistics in Medicine, 29, 9, (1037-1057), (2010). DOI: https://doi.org/10.1002/sim.3841 Durrleman, S, and Simon, R. Flexible regression models with cubic splines. Statistics in Medicine, 8, 5, (551-561), (1989). DOi: https://doi.org/10.1002/sim.4780080504

Akaike, Hirotugu (1974), «A new look at the statistical model identification», IEEE Transactions on Automatic Control 19 (6): 716-723, MR 0423716, doi:10.1109/TAC.1974.1100705.

The R foundation. Available on https://www.r-project.org/