NVD IEVA FLPP 2024

Modified

April 26, 2024

Packages

Set theme

Load the data

Remove P02 and n25 from

Recode Sex_ID

Eliminate Vecums, and change vecums_2 to Age

Add the MONTH

Merge 18 with 17 years old

EDA

NAs

Explore the patter of missingness

EDA

[1] 2887699      13

How many unique patients?

x
178810

In Latvia are 356.864 children 0 to 17 years-old

Source: https://data.stat.gov.lv/pxweb/lv/OSP_PUB/START__POP__IR__IRD/IRD041

So, this represents

[1] 50.10592

How many girls and boys

Age

SEX_ID	n	percent
Female	1465929	50.8%
Male	1421770	49.2%

How many specialist attentions PER patient?

SPEC_KODS	visits_count
Zobārsts	129215
Zobu higiēnists	111225
Radiologs	10517
Ortodonts	5180
Bērnu zobārsts	4150
Anesteziologs	3454
Mutes, sejas un žokļu ķirurgs	1380
Radiologa asistents	1367
Zobārsta asistents	342
Zobu tehniķis	159
Periodontologs	78

FINAL ANALYSIS

ZOBARSTS AND BERNU ZOBARSTS

Filter only Zobarsts and Bernu Z

Zobārsts Bērnu zobārsts

1. How many children visits 1, 2, or more times

Visits	n
1	59437
2	29292
3	16797
4	10015
5	5906
6	3554
7	2146
8	1324
9	839
> 9 visits	1480

Visits	n
1	59437
2	29292
3	16797
4	10015
> 4 visits	15249

2. How many children per age?

AGE	n	percent
0	41	0.0%
1	855	0.7%
2	2302	1.8%
3	3490	2.7%
4	5584	4.3%
5	8220	6.3%
6	10000	7.6%
7	12532	9.6%
8	10835	8.3%
9	10096	7.7%
10	9374	7.2%
11	8091	6.2%
12	7886	6.0%
13	8122	6.2%
14	8635	6.6%
15	8298	6.3%
16	7586	5.8%
17	8843	6.8%

This is interesting, seems children attend mostly in three ages.

3. How many times children goes to dentists in different age groups?

Now filter the unique PAC_ID

AGE	1	More than 1 time
0	39	2
1	784	71
2	2012	290
3	2604	886
4	3243	2341
5	3847	4373
6	4135	5865
7	5345	7187
8	4395	6440
9	4348	5748
10	4281	5093
11	3845	4246
12	3799	4087
13	3696	4426
14	3747	4888
15	3469	4829
16	3063	4523
17	2785	6058

4. How many patients per clinic?

AI_KODS	n
Less than 500 patients	38454
010019111	8007
010020301	7186
010064120	6063
010064521	4254
050024301	4224
010001535	4191
130000045	3147
170020401	2339
010064514	2153
210020301	1994
801000009	1977
210000058	1762
420200053	1740
170064506	1735
270064503	1673
620200014	1588
740200016	1576
400200025	1509
010064301	1459
210000043	1318
900200089	1317
740200059	1246
019277203	1127
270064004	1099
010000343	1090
090000093	1053
010054211	1024
019564503	1009
800600020	993
840200048	992
130024102	976
741400017	955
620200062	925
010064502	917
741400025	915
801600085	899
807600014	838
840200026	835
010001411	800
600200001	790
740200008	757
002000005	712
270000064	699
360800006	698
010001818	689
019164506	663
019464501	663
010064522	607
010001666	595
110000048	593
761200024	572
006000002	568
801200002	553
050000127	550
400200052	547
761200011	547
170077201	545
700200034	537
880200029	525
001000035	519
901200021	502

4.1. Visits per clinic

PAC_ID = Patient

MAN_DAT = Date of the visit

MP_CODE = Manipulations

AI_KODS	Total_Visits
010019111	14750
010020301	14692
010064120	13672
050024301	10146
010064521	9911
130000045	8972
010001535	8680
210020301	5968
010064514	5216
170020401	5046
270064503	4985
801000009	4783
270064004	4696
170064506	4208
210000058	3679
620200014	3679
420200053	3671
210000043	3279
900200089	3203
740200016	3108
740200059	3103
800600020	2958
380200022	2943
741400025	2934
400200025	2855
010000343	2649
010001881	2561
010064502	2438
019277203	2434
001000035	2276
019464501	2244
010064301	2224
270000064	2213
010054211	2177
090000093	2106
801600085	2106
019564503	2059
010001411	2043
010001818	2014
806077202	2014
840200048	2013
840200026	1973
006000002	1927
741400017	1840
600200001	1780
010064522	1677
130024102	1655
740200008	1642
360800006	1593
620200062	1583
801400003	1566
010001581	1488
380200011	1471
002000005	1357
807600014	1344
019164506	1338
681800001	1319
400200052	1272
110000048	1256
801200002	1170
130064502	1156
901200021	1152
880200029	1115
806900003	1107
800800006	1084
010000427	1044
090000016	1035
900200086	1029
010001666	1027
500200032	1014
700200034	1000
500200006	994
050000127	983
170077201	974
761200024	962
761200011	950
010044004	916
885100007	894
381600004	885
380200037	884
840200079	877
887600002	871
801600087	869
010001873	859
010001714	850
460800005	846
090020301	844
130000014	835
270000014	834
090000079	804
941600013	772
010077210	763
010064545	762
019164502	751
321400001	747
540200007	747
740200001	738
006000005	737
010001914	732
500200010	723
601000004	719
500200063	712
801000013	701
001000031	681
010000832	680
540200010	677
900200056	661
620200018	659
660200027	658
961600005	656
250000169	652
270000024	650
130000098	630
701400004	627
800600008	621
019364501	616
420200056	612
010001913	608
110000023	608
801800014	598
940200032	595
019464004	593
019464507	591
460200019	586
019577201	577
381600005	576
809600010	560
561800004	558
700200067	556
090024101	554
424700004	552
090000113	547
684900003	545
661400003	544
840200025	542
360200066	530
600200018	529
320200002	528
010000104	525
130064002	523
620200008	522
010064111	519
409500011	518
540200004	506
010000178	498
680200006	494
250000030	492
680200023	481
681000002	480
010001157	479
406477201	475
010001409	471
019164058	463
500200020	462
900200087	460
805200007	459
740200096	456
941600023	456
901200014	446
660200040	445
406464501	443
250000041	439
328277201	436
440200002	435
010064544	432
090077206	430
090077202	423
901200007	421
460200040	415
641000005	414
170000154	410
421200007	401
700800012	397
010001978	392
010064013	384
420200080	370
440200012	364
641000007	364
429300002	363
010001054	353
090065208	341
420200059	339
010001043	338
110000003	338
170000197	335
940200018	330
967300003	325
387500004	321
381600009	315
460800012	314
320200023	311
800600019	311
023000003	302
010000319	298
940200007	298
320200009	297
460200002	297
640600015	297
250000042	294
010001180	293
250000040	290
880200055	284
270000003	275
900200088	271
326100006	267
640600008	267
880200080	267
270064507	253
641000001	252
680200007	248
420200030	241
681000014	239
010000095	238
740600009	235
400200032	234
028000001	233
760200015	232
320200041	225
801600054	223
320200037	222
010065217	215
010001408	206
905700001	203
460200039	200
429300010	195
001000008	194
740600007	191
460200038	186
010001897	181
807600024	175
019677203	171
760200019	165
807477201	164
250000045	163
661400007	161
740200015	153
621200004	150
880200036	147
250000046	142
010054114	140
326100005	140
780200012	139
940200019	135
940200006	124
050000017	110
741000006	109
500200012	107
010077222	106
741400016	105
900200082	103
961000003	93
020000002	85
051000003	78
808477201	76
460200023	75
900200058	68
468900002	67
888300008	65
010000232	62
807635202	61
019464518	46
170000046	45
420200068	45
090000057	39
880200056	38
680200030	34
010011803	33
328200003	20
NA	3

5. How many patients per region?

PAC_ATVK	n
Less than 500 patients	40279
10093	10861
10094	10852
10092	7303
10095	6119
2000	5374
5000	4669
10096	4018
3000	3860
6000	3098
4000	3007
7000	2602
40010	2093
39410	1571
45200	1346
31010	1337
44420	1295
10091	1293
52210	1208
34420	1182
54010	1118
26200	1111
39200	1100
48200	1059
NA	1053
41400	973
39400	882
25200	857
46210	853
41200	781
40200	715
23400	696
33200	680
44400	635
34210	621
23410	611
40220	606
37210	526
24200	518
36200	515
51220	510
44410	503
29200	500

6. Why are the children coming?

IEMESLS	n
Regulâru apskati	1530903
NA	178893
Akûtâm sâpçm	167247
Traumu	14322

7. Extract the month from MAN_DAT

8. Reasons of visit per month

IEMESLS	1	2	3	4	5	6	7	8	9	10	11	12
Regulâru apskati	19008	17999	23556	20277	22314	21213	19081	23397	22967	23318	24165	19270
Akûtâm sâpçm	1960	1922	2167	1978	2205	1994	1946	2476	2559	2352	2390	2230
NA	1649	1746	2403	1953	2353	2360	2071	2853	2485	2533	2716	2285
Traumu	133	138	193	169	174	160	187	183	217	190	171	157

REMOVE OBJECT zobarst

HYGIENISTS

Filter only Zobu higiēnists

1. How many children visits 1, 2, or more times

Visits	n
1	98086
2	12210
3	857
4	70
5	1
6	1

Visits	n
1	98086
2	12210
3	857
4	70
> 4 visits	2

2. How many children per age?

AGE	n	percent
1	124	0.1%
2	1755	1.6%
3	4256	3.8%
4	5448	4.9%
5	6468	5.8%
6	7073	6.4%
7	8908	8.0%
8	9029	8.1%
9	7993	7.2%
10	7692	6.9%
11	7459	6.7%
12	7812	7.0%
13	8286	7.4%
14	8114	7.3%
15	7535	6.8%
16	6490	5.8%
17	6783	6.1%

This is interesting, seems children attend mostly in three ages.

3. How many times children goes to dentists in different age groups?

Now filter the unique PAC_ID

AGE	1	More than 1 time
1	123	1
2	1716	39
3	4039	217
4	5186	262
5	6112	356
6	6662	411
7	7329	1579
8	7188	1841
9	7290	703
10	7026	666
11	6149	1310
12	5936	1876
13	6807	1479
14	7408	706
15	6918	617
16	6029	461
17	6168	615

4. How many patients per clinic?

AI_KODS	n
Less than 500 patients	22497
420200053	8016
010019111	6876
010020301	6654
010064120	5351
050024301	3354
740200059	3325
010064521	3251
010001535	3183
010064514	2968
130000045	2857
210000058	2668
620200014	2021
090000057	1894
170020401	1805
210000043	1770
250000169	1718
010064301	1576
170064506	1555
400200025	1465
761200024	1354
019277203	1319
270064503	1195
090024001	1172
019564503	1111
019164506	1103
807600014	1062
801600085	1016
028000001	1007
840200048	960
805200007	948
010054211	934
760200003	846
900200089	845
620200062	804
010000343	759
800600020	713
320200041	694
880200029	690
010001873	689
010064502	671
010001818	657
741400025	638
801400003	626
840200026	623
130024102	614
900200006	610
010000832	603
460200002	582
400200052	548
170077202	513
740200016	511
NA	4

4.1. Visits per clinic

PAC_ID = Patient

MAN_DAT = Date of the visit

MP_CODE = Manipulations

AI_KODS	Total_Visits
420200053	8703
010019111	7756
010020301	7343
010064120	6156
740200059	4026
010064521	3943
050024301	3642
010001535	3459
010064514	3330
210000058	3143
130000045	3139
620200014	2316
090000057	2273
210000043	2027
170020401	1954
250000169	1844
400200025	1757
170064506	1724
010064301	1701
090024001	1554
019277203	1473
761200024	1465
270064503	1281
807600014	1240
019564503	1237
019164506	1197
801600085	1182
840200048	1102
010054211	1095
028000001	1090
805200007	1035
900200089	999
620200062	922
760200003	920
010000343	825
741400025	787
800600020	778
010001873	765
801400003	740
320200041	738
880200029	738
010064502	726
010001818	712
460200002	703
840200026	669
010000832	658
900200006	656
130024102	641
641000022	638
400200052	624
806900003	619
740200016	587
019464507	576
661400001	558
170077202	556
090000016	529
660200027	523
170077201	512
360200066	512
010044004	507
210020301	505
360800006	476
170000197	475
620200018	471
050000127	454
680200030	452
090077202	442
270000064	432
010064522	423
800800006	419
700200067	414
901200007	405
801600087	394
741400017	391
019364501	377
961600005	350
840200079	348
420200080	341
010001978	340
600200001	329
801800014	328
420200059	324
740200008	323
010077210	321
019464004	321
010064111	315
010001913	309
130000098	289
001000031	288
941600023	286
801200002	283
130064502	281
641600001	278
090024101	273
019164502	269
090065208	269
010000319	262
010001714	262
010001914	248
420200030	247
800600019	247
020000002	239
801000013	239
010001408	238
010064545	236
010001897	198
270064004	197
500200063	196
010064013	191
010001411	189
090000093	187
010001054	181
320200002	180
010001581	177
010001180	168
010064544	168
270000014	167
006000005	164
010001409	160
051000003	158
326100006	156
010000427	154
010001157	148
740600007	138
800600008	138
621200004	132
700800012	132
429300010	126
010000095	124
010065217	121
250000041	120
880200080	97
888300008	91
840200025	88
010001043	86
328200003	72
940200019	72
680200006	70
019164058	68
328277201	68
740200001	63
001000008	61
170000046	56
905700001	53
090077206	42
420200068	37
002000005	34
010054114	25
601000001	11
NA	4

5. How many patients per region?

PAC_ATVK	n
Less than 500 patients	35185
10094	10089
10093	8710
10092	6316
10095	5592
5000	4576
3000	3467
2000	3465
10096	3352
6000	2834
40010	2340
4000	1969
39410	1666
54010	1610
7000	1364
10091	1353
48200	1331
26200	1196
39200	1168
44420	1067
52210	1037
34420	990
45200	969
39400	945
25200	844
23400	786
NA	784
33200	724
44400	718
46210	706
23410	677
31010	623
40200	603
34210	593
28210	565
36200	507
41400	504

8. Reasons of visit per month

IEMESLS	1	2	3	4	5	6	7	8	9	10	11	12
Regulâru apskati	7280	6903	10201	8716	9975	8550	7095	10406	9720	10248	10276	8207
NA	708	757	964	863	948	1057	956	1447	1132	1295	1047	805
Akûtâm sâpçm	268	335	505	444	380	635	467	672	578	524	543	451
Traumu	1	2	4	2	NA	3	NA	3	3	1	1	1

REMOVE OBJECT hygienist

ALL 8

8. Reasons of visit per month

IEMESLS	1	2	3	4	5	6	7	8	9	10	11	12
Regulâru apskati	27326	25841	34524	30038	32890	30732	27175	34888	33283	34478	35122	28253
Akûtâm sâpçm	2368	2403	2879	2549	2713	2813	2579	3286	3275	3007	3066	2813
NA	2350	2459	3328	2791	3273	3357	2934	4169	3513	3735	3716	2990
Traumu	138	141	198	172	176	165	189	187	222	192	172	158

2024 April 11

Las preguntas siguentes de datos:

1. Verificamos, si no hemos dejado muchos datos sin analyzar:

Cuantas filas no tienen datos de SPEC_KODS?

# A tibble: 1 × 1
  NA_Count
     <int>
1        0

1. Verifikamos, cuantas visitas hay - cuantas veces esta el codigo 70001 (variable MP_CODE)

[1] 447849

--- title: "NVD IEVA FLPP 2024" date-modified: last-modified format: html: toc: true toc-expand: 3 code-fold: true code-tools: true editor: visual execute: echo: false cache: false warning: false message: false --- # Packages ```{r} # Load required libraries with pacman; installs them if not already installed pacman::p_load(tidyverse, # tools for data science scales, readr, naniar, #NAs janitor, # for data cleaning and tables here, # for reproducible research gtsummary # for tables ) ``` Set theme ```{r} theme_set(theme_minimal()) ``` # Load the data ```{r} MAVITO_ZPN_MAN_UD1_B <- read_rds(here("analysis", "data", "MAVITO_ZPN_MAN_UD1_B.rds")) |> filter(!is.na(SEX_ID)) ``` ### Remove P02 and n25 from ```{r} MAVITO_ZPN_MAN_UD1_B <- MAVITO_ZPN_MAN_UD1_B |> filter(!str_detect(SPEC_KODS, c("P02|n25") )) ``` ### Recode Sex_ID ```{r} MAVITO_ZPN_MAN_UD1_B <- MAVITO_ZPN_MAN_UD1_B |> mutate(SEX_ID = recode(SEX_ID, "1" = "Male", "2" = "Female")) ``` ### Eliminate Vecums, and change vecums_2 to Age ```{r} MAVITO_ZPN_MAN_UD1_B <- MAVITO_ZPN_MAN_UD1_B |> select(-VECUMS, "AGE" = "VECUMS_2") ``` ### Add the MONTH ```{r} MAVITO_ZPN_MAN_UD1_B <- MAVITO_ZPN_MAN_UD1_B |> mutate(MONTH = lubridate::month(MAN_DAT) ) |> relocate(MONTH, .after = MAN_DAT) ``` Merge 18 with 17 years old ```{r} MAVITO_ZPN_MAN_UD1_B <- MAVITO_ZPN_MAN_UD1_B %>% mutate(AGE = ifelse(AGE > 17, 17, AGE)) ``` # EDA ## NAs ```{r} MAVITO_ZPN_MAN_UD1_B |> naniar::gg_miss_var(show_pct = TRUE) + labs(title = "Missing data") ``` Explore the patter of missingness ```{r} # gg_miss_upset(MAVITO_ZPN_MAN_UD1_B) ``` ## EDA ```{r} dim(MAVITO_ZPN_MAN_UD1_B) ``` ### How many unique patients? ```{r} n_distinct(MAVITO_ZPN_MAN_UD1_B$PAC_ID) |> knitr::kable() ``` In Latvia are 356.864 children 0 to 17 years-old Source: <https://data.stat.gov.lv/pxweb/lv/OSP_PUB/START__POP__IR__IRD/IRD041> So, this represents ```{r} n_distinct(MAVITO_ZPN_MAN_UD1_B$PAC_ID) / 356864 *100 ``` ## How many girls and boys ## Age ```{r} MAVITO_ZPN_MAN_UD1_B |> tabyl(SEX_ID) |> adorn_pct_formatting() |> knitr::kable() ``` ```{r} MAVITO_ZPN_MAN_UD1_B |> ggplot(aes(x = AGE)) + geom_histogram(bins = 10) + labs(title = "Age Distribution") ``` ```{r} # Calculate median ages for each SEX_ID median_ages <- MAVITO_ZPN_MAN_UD1_B |> group_by(SEX_ID) |> summarise(median_age = median(AGE, na.rm = TRUE)) # Plotting MAVITO_ZPN_MAN_UD1_B |> filter(!is.na(SEX_ID)) |> ggplot(aes(x = AGE)) + geom_histogram(bins = 10) + facet_grid(SEX_ID ~ .) + geom_vline(data = median_ages, aes(xintercept = median_age, color = as.factor(SEX_ID)), linetype = "dashed") + scale_color_discrete(name = "SEX_ID Median") + theme_minimal() + labs(x = "Age", y = "Count", title = "Histogram of Ages by SEX_ID with Median Lines") ``` ```{r} rm(median_ages) ``` ### How many specialist attentions PER patient? ```{r} MAVITO_ZPN_MAN_UD1_B |> group_by(SPEC_KODS) |> summarise(visits_count = n_distinct(PAC_ID)) |> arrange(desc(visits_count)) |> knitr::kable() ``` ```{r} MAVITO_ZPN_MAN_UD1_B |> filter(!is.na(SPEC_KODS)) |> group_by(SPEC_KODS) |> summarise(visits_count = n_distinct(PAC_ID)) |> arrange(desc(visits_count)) |> ggplot(aes(x = fct_reorder(SPEC_KODS,visits_count), y = visits_count)) + geom_col() + coord_flip() + scale_y_continuous(labels = label_number()) + labs(title = "Visits by specialist", x = "", y = "Visits") ``` # FINAL ANALYSIS ## ZOBARSTS AND BERNU ZOBARSTS ### Filter only Zobarsts and Bernu Z ``` Zobārsts Bērnu zobārsts ``` ```{r} MAVITO_ZPN_MAN_UD1_B_zobarst <- MAVITO_ZPN_MAN_UD1_B |> filter(str_detect(SPEC_KODS, "Zobārsts|Bērnu zobārsts")) ``` ### 1. How many children visits 1, 2, or more times ```{r} # unique visits MAVITO_ZPN_MAN_UD1_B_zobarst |> distinct(PAC_ID, MAN_DAT) |> group_by(PAC_ID) |> summarise(Visits = n()) |> ungroup() |> # how many PAC_IDs had exactly 1 visit, 2 visits, etc. group_by(Visits) |> summarise(n = n()) |> # plot ggplot(aes(x = Visits, y = n)) + geom_col() ``` ```{r} MAVITO_ZPN_MAN_UD1_B_zobarst |> distinct(PAC_ID, MAN_DAT) |> group_by(PAC_ID) |> group_by(PAC_ID) |> summarise(Visits = n()) |> # how many PAC_IDs had exactly 1 visit, 2 visits, etc. group_by(Visits) |> summarise(n = n()) |> mutate(Visits = if_else(Visits > 9, "> 9 visits", as.character(Visits))) |> group_by(Visits) |> summarise(n = sum(n)) |> knitr::kable() ``` ```{r} MAVITO_ZPN_MAN_UD1_B_zobarst |> distinct(PAC_ID, MAN_DAT) |> group_by(PAC_ID) |> group_by(PAC_ID) |> summarise(Visits = n()) |> # how many PAC_IDs had exactly 1 visit, 2 visits, etc. group_by(Visits) |> summarise(n = n()) |> mutate(Visits = if_else(Visits > 4, "> 4 visits", as.character(Visits))) |> group_by(Visits) |> summarise(n = sum(n)) |> knitr::kable() ``` ### 2. How many children per age? ```{r} MAVITO_ZPN_MAN_UD1_B_zobarst |> distinct(PAC_ID, .keep_all = T) |> group_by(AGE) |> tabyl(AGE) |> adorn_pct_formatting() |> knitr::kable() ``` ```{r} MAVITO_ZPN_MAN_UD1_B_zobarst |> distinct(PAC_ID, .keep_all = T) |> ggplot(aes(x = AGE)) + geom_histogram(bins = 16) + labs(title = "Age (unique patients)", y = "N") ``` This is interesting, seems children attend mostly in three ages. ### 3. How many times children goes to dentists in different age groups? ```{r} # Count the number of visits for each child child_visit_counts <- MAVITO_ZPN_MAN_UD1_B_zobarst |> distinct(PAC_ID, MAN_DAT) |> group_by(PAC_ID) |> summarise(Num_Visits = n(), .groups = 'drop') # Count visits ``` ```{r} # join the visit counts with the MAVITO_ZPN_MAN_UD1_B_zobarst dataset MAVITO_ZPN_MAN_UD1_B_zobarst <- MAVITO_ZPN_MAN_UD1_B_zobarst |> left_join(child_visit_counts, by = "PAC_ID") ``` ```{r} rm(child_visit_counts) ``` Now filter the unique PAC_ID ```{r} MAVITO_ZPN_MAN_UD1_B_zobarst |> distinct(PAC_ID, .keep_all = T) |> mutate(Num_Visits = if_else(Num_Visits > 1, "More than 1 time", as.character(Num_Visits))) |> tabyl(AGE, Num_Visits) |> knitr::kable() ``` ```{r} MAVITO_ZPN_MAN_UD1_B_zobarst |> distinct(PAC_ID, .keep_all = T) |> mutate(Num_Visits = if_else(Num_Visits > 1, "Multiple visits", as.character("One visit"))) |> tabyl(AGE, Num_Visits) |> pivot_longer(!AGE, names_to = "Group", values_to = "n") |> ggplot( aes(x = AGE, y = n, fill = Group)) + geom_bar(stat = "identity", position = "identity", alpha = 0.5) + labs(title = "Distribution of Visits by Age", x = "Age", y = "Number of Children - Dentist", fill = "Visits") ``` ### 4. How many patients per clinic? ```{r} MAVITO_ZPN_MAN_UD1_B_zobarst |> distinct(PAC_ID, .keep_all = TRUE) |> # not included repeated visits mutate(AI_KODS = as.character(AI_KODS), # Convert AI_KODS to character AI_KODS = fct_lump_min(AI_KODS, min = 500, other_level = "Less than 500 patients")) |> group_by(AI_KODS) |> count() |> arrange(desc(n)) |> knitr::kable() ``` ```{r} MAVITO_ZPN_MAN_UD1_B_zobarst |> distinct(PAC_ID, .keep_all = TRUE) |> # not included repeated visits mutate(AI_KODS = as.character(AI_KODS), # Convert AI_KODS to character AI_KODS = fct_lump_min(AI_KODS, min = 500, other_level = "Less than 500 patients")) |> group_by(AI_KODS) |> count() |> ggplot(aes(x = fct_reorder(AI_KODS, n), y = n)) + geom_col() + coord_flip() + labs(title = "Patients per Clinic", x = "Clinic", y = "Patients") ``` ```{r} MAVITO_ZPN_MAN_UD1_B_zobarst |> distinct(PAC_ID, .keep_all = TRUE) |> # not included repeated visits # mutate(AI_KODS = as.character(AI_KODS), # Convert AI_KODS to character # AI_KODS = fct_lump_min(AI_KODS, min = 500, other_level = "Less than 500 patients")) |> group_by(AI_KODS) |> count() |> arrange(desc(n)) |> write_csv(here("data", "_zobarsts_patientes_per_clinic.csv")) ``` ### 4.1. Visits per clinic PAC_ID = Patient MAN_DAT = Date of the visit MP_CODE = Manipulations ```{r} MAVITO_ZPN_MAN_UD1_B_zobarst |> group_by(AI_KODS, PAC_ID) |> # Group by clinic and patient summarise(Visits = n_distinct(MAN_DAT)) |> # Count unique visit dates per patient per clinic group_by(AI_KODS) |> # Group by clinic summarise(Total_Visits = sum(Visits)) |> # Sum visits for each clinic arrange(desc(Total_Visits)) |> knitr::kable() ``` ```{r} MAVITO_ZPN_MAN_UD1_B_zobarst |> group_by(AI_KODS, PAC_ID) |> # Group by clinic and patient summarise(Visits = n_distinct(MAN_DAT)) |> # Count unique visit dates per patient per clinic group_by(AI_KODS) |> # Group by clinic summarise(Total_Visits = sum(Visits)) |> # Sum visits for each clinic arrange(desc(Total_Visits)) |> ggplot(aes(x = fct_reorder(AI_KODS, Total_Visits), y = Total_Visits)) + geom_col() + coord_flip() + theme(legend.position = "none") + labs(title = "Visits per clinic", x = "Clinic") ``` ### 5. How many patients per region? ```{r} MAVITO_ZPN_MAN_UD1_B_zobarst |> distinct(PAC_ID, .keep_all = TRUE) |> # not included repeated visits mutate(PAC_ATVK = as.character(PAC_ATVK), # Convert PAC_ATVK to character PAC_ATVK = fct_lump_min(PAC_ATVK, min = 500, other_level = "Less than 500 patients")) |> group_by(PAC_ATVK) |> count() |> arrange(desc(n)) |> knitr::kable() ``` ```{r} MAVITO_ZPN_MAN_UD1_B_zobarst |> distinct(PAC_ID, .keep_all = TRUE) |> # not included repeated visits mutate(PAC_ATVK = as.character(PAC_ATVK), # Convert PAC_ATVK to character PAC_ATVK = fct_lump_min(PAC_ATVK, min = 500, other_level = "Less than 500 patients")) |> group_by(PAC_ATVK) |> count() |> ggplot(aes(x = fct_reorder(PAC_ATVK, n), y = n)) + geom_col() + coord_flip() + labs(title = "Patients per Region", x = "Region", y = "Patients") ``` ```{r} # Save the ungrouped data to an CSV for Ilze MAVITO_ZPN_MAN_UD1_B_zobarst |> distinct(PAC_ID, .keep_all = TRUE) |> # not included repeated visits mutate(PAC_ATVK = as.character(PAC_ATVK)) |> # Convert PAC_ATVK to character group_by(PAC_ATVK) |> count() |> arrange(desc(n)) |> write_csv(here("data", "zobarsts_patientes_per_clinic.csv")) ``` ### 6. Why are the children coming? ```{r} MAVITO_ZPN_MAN_UD1_B_zobarst |> group_by(IEMESLS) |> count() |> arrange(desc(n)) |> knitr::kable() ``` ```{r} MAVITO_ZPN_MAN_UD1_B_zobarst |> group_by(IEMESLS) |> count() |> arrange(desc(n)) |> ggplot(aes(x = fct_reorder(IEMESLS, n, .desc = T) , y = n)) + geom_col() + labs(title = "What are the reasons for children's dental visits?", x = "", y = "Visits") ``` ### 7. Extract the month from MAN_DAT ```{r} # MAVITO_ZPN_MAN_UD1_B_zobarst <- MAVITO_ZPN_MAN_UD1_B_zobarst |> # mutate(MONTH = lubridate::month(MAN_DAT) ) |> # relocate(MONTH, .after = MAN_DAT) ``` ### 8. Reasons of visit per month ```{r} MAVITO_ZPN_MAN_UD1_B_zobarst |> group_by(MONTH, IEMESLS) |> # Group by month and reason for the visit summarise(Visits = n_distinct(PAC_ID, MAN_DAT)) |> # Count unique patient visits (by patient and date) for each reason arrange(MONTH, desc(Visits)) |> # Arrange by month and number of visits pivot_wider(names_from = MONTH, values_from = Visits) |> knitr::kable() ``` ```{r} MAVITO_ZPN_MAN_UD1_B_zobarst |> group_by(MONTH, IEMESLS) |> # Group by month and reason for the visit summarise(Visits = n_distinct(PAC_ID, MAN_DAT)) |> # Count unique patient visits (by patient and date) for each reason ggplot(aes(x = as.factor(MONTH), y = Visits, color = IEMESLS, group = IEMESLS)) + geom_line() + scale_y_log10() + labs(title = "Reasons of visits per month - Dentists", y = "Visits (log10)", x = "Month") + theme(legend.position = "top") ``` ### REMOVE OBJECT zobarst ```{r} rm(MAVITO_ZPN_MAN_UD1_B_zobarst) ``` ## HYGIENISTS ### Filter only Zobu higiēnists ```{r} MAVITO_ZPN_MAN_UD1_B_hygienist <- MAVITO_ZPN_MAN_UD1_B |> filter(str_detect(SPEC_KODS, "Zobu higiēnists")) ``` ### 1. How many children visits 1, 2, or more times ```{r} # unique visits MAVITO_ZPN_MAN_UD1_B_hygienist |> distinct(PAC_ID, MAN_DAT) |> group_by(PAC_ID) |> summarise(Visits = n()) |> ungroup() |> # how many PAC_IDs had exactly 1 visit, 2 visits, etc. group_by(Visits) |> summarise(n = n()) |> # plot ggplot(aes(x = Visits, y = n)) + geom_col() ``` ```{r} MAVITO_ZPN_MAN_UD1_B_hygienist |> distinct(PAC_ID, MAN_DAT) |> group_by(PAC_ID) |> group_by(PAC_ID) |> summarise(Visits = n()) |> # how many PAC_IDs had exactly 1 visit, 2 visits, etc. group_by(Visits) |> summarise(n = n()) |> mutate(Visits = if_else(Visits > 9, "> 9 visits", as.character(Visits))) |> group_by(Visits) |> summarise(n = sum(n)) |> knitr::kable() ``` ```{r} MAVITO_ZPN_MAN_UD1_B_hygienist |> distinct(PAC_ID, MAN_DAT) |> group_by(PAC_ID) |> group_by(PAC_ID) |> summarise(Visits = n()) |> # how many PAC_IDs had exactly 1 visit, 2 visits, etc. group_by(Visits) |> summarise(n = n()) |> mutate(Visits = if_else(Visits > 4, "> 4 visits", as.character(Visits))) |> group_by(Visits) |> summarise(n = sum(n)) |> knitr::kable() ``` ### 2. How many children per age? ```{r} MAVITO_ZPN_MAN_UD1_B_hygienist |> distinct(PAC_ID, .keep_all = T) |> group_by(AGE) |> tabyl(AGE) |> adorn_pct_formatting() |> knitr::kable() ``` ```{r} MAVITO_ZPN_MAN_UD1_B_hygienist |> distinct(PAC_ID, .keep_all = T) |> ggplot(aes(x = AGE)) + geom_histogram(bins = 12) + labs(title = "Age (unique patients)", y = "N") ``` This is interesting, seems children attend mostly in three ages. ### 3. How many times children goes to dentists in different age groups? ```{r} # Count the number of visits for each child child_visit_counts <- MAVITO_ZPN_MAN_UD1_B_hygienist |> distinct(PAC_ID, MAN_DAT) |> group_by(PAC_ID) |> summarise(Num_Visits = n(), .groups = 'drop') # Count visits ``` ```{r} # join the visit counts with the MAVITO_ZPN_MAN_UD1_B_hygienist dataset MAVITO_ZPN_MAN_UD1_B_hygienist <- MAVITO_ZPN_MAN_UD1_B_hygienist |> left_join(child_visit_counts, by = "PAC_ID") ``` ```{r} rm(child_visit_counts) ``` Now filter the unique PAC_ID ```{r} MAVITO_ZPN_MAN_UD1_B_hygienist |> distinct(PAC_ID, .keep_all = T) |> mutate(Num_Visits = if_else(Num_Visits > 1, "More than 1 time", as.character(Num_Visits))) |> tabyl(AGE, Num_Visits) |> knitr::kable() ``` ```{r} MAVITO_ZPN_MAN_UD1_B_hygienist |> distinct(PAC_ID, .keep_all = T) |> mutate(Num_Visits = if_else(Num_Visits > 1, "Multiple visits", as.character("One visit"))) |> tabyl(AGE, Num_Visits) |> pivot_longer(!AGE, names_to = "Group", values_to = "n") |> ggplot( aes(x = AGE, y = n, fill = Group)) + geom_bar(stat = "identity", position = "identity", alpha = 0.5) + labs(title = "Distribution of Visits by Age - Hygienist ", x = "Age", y = "Number of Children", fill = "Visits") ``` ### 4. How many patients per clinic? ```{r} MAVITO_ZPN_MAN_UD1_B_hygienist |> distinct(PAC_ID, .keep_all = TRUE) |> # not included repeated visits mutate(AI_KODS = as.character(AI_KODS), # Convert AI_KODS to character AI_KODS = fct_lump_min(AI_KODS, min = 500, other_level = "Less than 500 patients")) |> group_by(AI_KODS) |> count() |> arrange(desc(n)) |> knitr::kable() ``` ```{r} MAVITO_ZPN_MAN_UD1_B_hygienist |> distinct(PAC_ID, .keep_all = TRUE) |> # not included repeated visits mutate(AI_KODS = as.character(AI_KODS), # Convert AI_KODS to character AI_KODS = fct_lump_min(AI_KODS, min = 500, other_level = "Less than 500 patients")) |> group_by(AI_KODS) |> count() |> ggplot(aes(x = fct_reorder(AI_KODS, n), y = n)) + geom_col() + coord_flip() + labs(title = "Patients per Clinic", x = "Clinic", y = "Patients") ``` ### 4.1. Visits per clinic PAC_ID = Patient MAN_DAT = Date of the visit MP_CODE = Manipulations ```{r} MAVITO_ZPN_MAN_UD1_B_hygienist |> group_by(AI_KODS, PAC_ID) |> # Group by clinic and patient summarise(Visits = n_distinct(MAN_DAT)) |> # Count unique visit dates per patient per clinic group_by(AI_KODS) |> # Group by clinic summarise(Total_Visits = sum(Visits)) |> # Sum visits for each clinic arrange(desc(Total_Visits)) |> knitr::kable() ``` ```{r} MAVITO_ZPN_MAN_UD1_B_hygienist |> group_by(AI_KODS, PAC_ID) |> # Group by clinic and patient summarise(Visits = n_distinct(MAN_DAT)) |> # Count unique visit dates per patient per clinic group_by(AI_KODS) |> # Group by clinic summarise(Total_Visits = sum(Visits)) |> # Sum visits for each clinic arrange(desc(Total_Visits)) |> ggplot(aes(x = fct_reorder(AI_KODS, Total_Visits), y = Total_Visits)) + geom_col() + coord_flip() + theme(legend.position = "none") + labs(title = "Visits per clinic", x = "Clinic") ``` ### 5. How many patients per region? ```{r} MAVITO_ZPN_MAN_UD1_B_hygienist |> distinct(PAC_ID, .keep_all = TRUE) |> # not included repeated visits mutate(PAC_ATVK = as.character(PAC_ATVK), # Convert PAC_ATVK to character PAC_ATVK = fct_lump_min(PAC_ATVK, min = 500, other_level = "Less than 500 patients")) |> group_by(PAC_ATVK) |> count() |> arrange(desc(n)) |> knitr::kable() ``` ```{r} MAVITO_ZPN_MAN_UD1_B_hygienist |> distinct(PAC_ID, .keep_all = TRUE) |> # not included repeated visits mutate(PAC_ATVK = as.character(PAC_ATVK), # Convert PAC_ATVK to character PAC_ATVK = fct_lump_min(PAC_ATVK, min = 500, other_level = "Less than 500 patients")) |> group_by(PAC_ATVK) |> count() |> ggplot(aes(x = fct_reorder(PAC_ATVK, n), y = n)) + geom_col() + coord_flip() + labs(title = "Patients per Region", x = "Region", y = "Patients") ``` ```{r} # Save the ungrouped data to an CSV for Ilze MAVITO_ZPN_MAN_UD1_B_hygienist |> distinct(PAC_ID, .keep_all = TRUE) |> # not included repeated visits mutate(PAC_ATVK = as.character(PAC_ATVK)) |> # Convert PAC_ATVK to character group_by(PAC_ATVK) |> count() |> arrange(desc(n)) |> write_csv(here("data", "hygienist_patientes_per_clinic.csv")) ``` ### 8. Reasons of visit per month ```{r} MAVITO_ZPN_MAN_UD1_B_hygienist |> group_by(MONTH, IEMESLS) |> # Group by month and reason for the visit summarise(Visits = n_distinct(PAC_ID, MAN_DAT)) |> # Count unique patient visits (by patient and date) for each reason arrange(MONTH, desc(Visits)) |> # Arrange by month and number of visits pivot_wider(names_from = MONTH, values_from = Visits) |> knitr::kable() ``` ```{r} MAVITO_ZPN_MAN_UD1_B_hygienist |> group_by(MONTH, IEMESLS) |> # Group by month and reason for the visit summarise(Visits = n_distinct(PAC_ID, MAN_DAT)) |> # Count unique patient visits (by patient and date) for each reason ggplot(aes(x = as.factor(MONTH), y = Visits, color = IEMESLS, group = IEMESLS)) + geom_line() + scale_y_log10() + labs(title = "Reasons of visits per month - Hygienist", y = "Visits (log10)", x = "Month") + theme(legend.position = "top") ``` ### REMOVE OBJECT hygienist ```{r} rm(MAVITO_ZPN_MAN_UD1_B_hygienist) ``` ## ALL 8 ### 8. Reasons of visit per month ```{r} MAVITO_ZPN_MAN_UD1_B |> group_by(MONTH, IEMESLS) |> # Group by month and reason for the visit summarise(Visits = n_distinct(PAC_ID, MAN_DAT)) |> # Count unique patient visits (by patient and date) for each reason arrange(MONTH, desc(Visits)) |> # Arrange by month and number of visits pivot_wider(names_from = MONTH, values_from = Visits) |> knitr::kable() ``` ```{r} MAVITO_ZPN_MAN_UD1_B |> group_by(MONTH, IEMESLS) |> # Group by month and reason for the visit summarise(Visits = n_distinct(PAC_ID, MAN_DAT)) |> # Count unique patient visits (by patient and date) for each reason ggplot(aes(x = as.factor(MONTH), y = Visits, color = IEMESLS, group = IEMESLS)) + geom_line() + scale_y_log10() + labs(title = "Reasons of visits per month - All", y = "Visits (log10)", x = "Month") + theme(legend.position = "top") ``` ## 2024 April 11 Las preguntas siguentes de datos: ### 1. Verificamos, si no hemos dejado muchos datos sin analyzar: Cuantas filas no tienen datos de SPEC_KODS? ```{r} MAVITO_ZPN_MAN_UD1_B %>% summarise(NA_Count = sum(is.na(SPEC_KODS))) # Count NA values in SPEC_KODS ``` ### 1. Verifikamos, cuantas visitas hay - cuantas veces esta el codigo 70001 (variable MP_CODE) ```{r} MAVITO_ZPN_MAN_UD1_B %>% filter(MP_CODE == 70001) %>% # Filter cells with code 70001 nrow() # Count rows ```