A Global Eye Bank Dataset into Trusted Intelligence Using SAS and R

Vision Without Validation Becomes Blind Analytics: Transforming a Global Eye Bank Dataset into Trusted Intelligence Using SAS and R

Introduction:From Corneal Donation Delays to Analytical Failure

Imagine a multinational ophthalmology research consortium preparing a global report on corneal transplant accessibility. The study combines operational records from eye banks across continents to evaluate donor availability, transplant waiting periods, and regional healthcare inequality.

Three weeks before executive presentation, statisticians discover a disaster.

Several eye banks appear twice under slightly different names. Some donor counts are negative because of data entry errors. Missing collection dates cause survival analyses to fail. Regional labels appear as APAC, Apac, Asia-Pacific, and ASIA PACIFIC simultaneously. A few email addresses contain spaces while others are malformed entirely. Some eye banks report impossible utilization rates exceeding 100%.

The dashboard shows misleading trends.

Machine learning models generate incorrect forecasts.

Funding decisions become unreliable.

Regulatory submissions become questionable.

This is not a software problem.

This is a data quality problem.

Experienced Clinical SAS Programmers understand that poor-quality source data can silently destroy statistical conclusions long before model building begins.

In modern healthcare analytics, data cleaning is not preprocessing.

It is risk management.

Global Eye Banks Dataset Overview

Our analytical dataset contains information collected from major eye banks worldwide.

Dataset Variables (9 Variables)

Variable	Description
EYE_BANK_ID	Unique eye bank identifier
EYE_BANK_NAME	Eye bank name
COUNTRY	Country location
REGION_CODE	Geographic region
DONORS_RECEIVED	Annual corneal donors
TRANSPLANTS_DONE	Annual transplants
AVG_WAIT_DAYS	Average waiting days
CONTACT_EMAIL	Operational email
LAST_AUDIT_DATE	Last regulatory audit date

Intentional corruption includes:

• Duplicate IDs
• Negative donor values
• Missing audit dates
• Invalid email addresses
• Incorrect region codes
• Mixed text case
• Leading spaces
• Impossible transplant counts
• NULL strings
• Invalid date values

1.SAS Raw Dataset Creation with Intentional Errors

data eyebank_raw;

infile datalines dsd dlm=',' truncover;

length eye_bank_id $8 eye_bank_name $45 country $25 region_code $20

       contact_email $60;

informat last_audit_date anydtdte20.;

format last_audit_date date9.;

input eye_bank_id :$8. eye_bank_name :$45. country :$25. region_code :$20.

      donors_received transplants_done avg_wait_days contact_email :$60.

      last_audit_date :anydtdte20.;

datalines;

EB001,LionsEyeBank,India,APAC,450,420,32,support@lions.org,12JAN2025

EB002,VisionTrust,USA,NA,620,590,28,info@visiontrust.org,15FEB2025

EB003,GlobalEye,UK,EUROPE,-45,310,20,admin@globaleye.org,11MAR2025

EB003,GlobalEye,UK,Europe,500,310,20,admin@globaleye.org,11MAR2025

EB004,CorneaCare,India,APAC,520,700,18,NULL,.

EB005,BrightSight,Canada,na,410,390,25,help@brightsight,14APR2025

EB006,FutureVision,Australia,APAC,300,280,22,futurevision.org,30FEB2025

EB007,EyeLife,Germany,EU,290,275,-5,contact@eyelife.org,19MAY2025

EB008,WorldCornea,India,AsiaPacific,610,605,31,world@cornea.org,.

EB009,ClearView,Japan,APAC,510,490,27,clear@view.org,10JUN2025

EB010,VisionTrust,USA,NA,620,590,28,info@visiontrust.org,15FEB2025

EB011,Eye Hope,Brazil,LATAM,400,410,30,hope @eye.org,01JUL2025

EB012,RetinaGift,India,APAC,.,340,40,retina@gift.org,20AUG2025

EB013,NewLight,UAE,MEA,370,355,29,newlight@mail.com,15SEP2025

EB014,OcularBank,France,eu,280,270,23,ocular@bank.org,15OCT2025

EB015,DonorVision,India,APAC,530,520,26,donor@vision.org,22NOV2025

EB016,LifeSight,Kenya,AFRICA,250,240,34,sight@life.org,11DEC2025

EB017,GiftVision,India,APAC,0,230,36,gift@vision.org,.

EB018,HopeCornea,USA,NorthAmerica,550,530,21,hope@cornea.org,15JAN2025

EB019,EyeFuture,India,APAC,460,455,24,eyefuture.com,14FEB2025

EB020,GlobalCornea,Singapore,APAC,390,370,19,global@cornea.org,15MAR2025

EB021,BrightVision,India,APAC,440,435,33,bright@vision.org,01APR2025

EB022,ClearGift,UK,EUROPE,480,470,27,cleargift@eye.org,18MAY2025

;

run;

proc print data=eyebank_raw;

run;

OUTPUT:

Obs	eye_bank_id	eye_bank_name	country	region_code	contact_email	last_audit_date	donors_received	transplants_done	avg_wait_days
1	EB001	LionsEyeBank	India	APAC	support@lions.org	12JAN2025	450	420	32
2	EB002	VisionTrust	USA	NA	info@visiontrust.org	15FEB2025	620	590	28
3	EB003	GlobalEye	UK	EUROPE	admin@globaleye.org	11MAR2025	-45	310	20
4	EB003	GlobalEye	UK	Europe	admin@globaleye.org	11MAR2025	500	310	20
5	EB004	CorneaCare	India	APAC	NULL	.	520	700	18
6	EB005	BrightSight	Canada	na	help@brightsight	14APR2025	410	390	25
7	EB006	FutureVision	Australia	APAC	futurevision.org	.	300	280	22
8	EB007	EyeLife	Germany	EU	contact@eyelife.org	19MAY2025	290	275	-5
9	EB008	WorldCornea	India	AsiaPacific	world@cornea.org	.	610	605	31
10	EB009	ClearView	Japan	APAC	clear@view.org	10JUN2025	510	490	27
11	EB010	VisionTrust	USA	NA	info@visiontrust.org	15FEB2025	620	590	28
12	EB011	Eye Hope	Brazil	LATAM	hope @eye.org	01JUL2025	400	410	30
13	EB012	RetinaGift	India	APAC	retina@gift.org	20AUG2025	.	340	40
14	EB013	NewLight	UAE	MEA	newlight@mail.com	15SEP2025	370	355	29
15	EB014	OcularBank	France	eu	ocular@bank.org	15OCT2025	280	270	23
16	EB015	DonorVision	India	APAC	donor@vision.org	22NOV2025	530	520	26
17	EB016	LifeSight	Kenya	AFRICA	sight@life.org	11DEC2025	250	240	34
18	EB017	GiftVision	India	APAC	gift@vision.org	.	0	230	36
19	EB018	HopeCornea	USA	NorthAmerica	hope@cornea.org	15JAN2025	550	530	21
20	EB019	EyeFuture	India	APAC	eyefuture.com	14FEB2025	460	455	24
21	EB020	GlobalCornea	Singapore	APAC	global@cornea.org	15MAR2025	390	370	19
22	EB021	BrightVision	India	APAC	bright@vision.org	01APR2025	440	435	33
23	EB022	ClearGift	UK	EUROPE	cleargift@eye.org	18MAY2025	480	470	27

Why LENGTH Must Come First

A common production mistake is defining LENGTH after assignment statements.

eye_bank_name="International Eye Donation Registry";

length eye_bank_name $20;

The variable becomes permanently truncated because SAS allocates memory during compilation.

This creates silent corruption.

R behaves differently because character vectors dynamically allocate memory rather than using fixed-length storage definitions.

Clinical production programmers always define LENGTH before assignment logic.

2.Cleaning Workflow Using DATA Step Programming

data eyebank_clean;

set eyebank_raw;

eye_bank_name=propcase(strip(eye_bank_name));

country=propcase(strip(country));

region_code=upcase(strip(region_code));

contact_email=compress(lowcase(contact_email),' ');

if donors_received <0 then donors_received=abs(donors_received);

if avg_wait_days<0 then avg_wait_days=.;

if missing(last_audit_date) then

last_audit_date=intnx('month',today(),-3,'same');

if index(contact_email,'@')=0 then

contact_email='unknown@placeholder.org';

if transplants_done>donors_received then

transplants_done=donors_received;

if region_code in ('EU','EUROPE') then region_code='EMEA';

select(region_code);

when('NA','NORTHAMERICA') region_code='NA';

when('APAC','ASIAPACIFIC') region_code='APAC';

otherwise;

end;

run;

proc print data=eyebank_clean;;

run;

OUTPUT:

Obs	eye_bank_id	eye_bank_name	country	region_code	contact_email	last_audit_date	donors_received	transplants_done	avg_wait_days
1	EB001	Lionseyebank	India	APAC	support@lions.org	12JAN2025	450	420	32
2	EB002	Visiontrust	Usa	NA	info@visiontrust.org	15FEB2025	620	590	28
3	EB003	Globaleye	Uk	EMEA	admin@globaleye.org	11MAR2025	45	45	20
4	EB003	Globaleye	Uk	EMEA	admin@globaleye.org	11MAR2025	500	310	20
5	EB004	Corneacare	India	APAC	unknown@placeholder.org	04APR2026	520	520	18
6	EB005	Brightsight	Canada	NA	help@brightsight	14APR2025	410	390	25
7	EB006	Futurevision	Australia	APAC	unknown@placeholder.org	04APR2026	300	280	22
8	EB007	Eyelife	Germany	EMEA	contact@eyelife.org	19MAY2025	290	275	.
9	EB008	Worldcornea	India	APAC	world@cornea.org	04APR2026	610	605	31
10	EB009	Clearview	Japan	APAC	clear@view.org	10JUN2025	510	490	27
11	EB010	Visiontrust	Usa	NA	info@visiontrust.org	15FEB2025	620	590	28
12	EB011	Eye Hope	Brazil	LATAM	hope@eye.org	01JUL2025	400	400	30
13	EB012	Retinagift	India	APAC	retina@gift.org	20AUG2025	.	.	40
14	EB013	Newlight	Uae	MEA	newlight@mail.com	15SEP2025	370	355	29
15	EB014	Ocularbank	France	EMEA	ocular@bank.org	15OCT2025	280	270	23
16	EB015	Donorvision	India	APAC	donor@vision.org	22NOV2025	530	520	26
17	EB016	Lifesight	Kenya	AFRICA	sight@life.org	11DEC2025	250	240	34
18	EB017	Giftvision	India	APAC	gift@vision.org	04APR2026	0	0	36
19	EB018	Hopecornea	Usa	NA	hope@cornea.org	15JAN2025	550	530	21
20	EB019	Eyefuture	India	APAC	unknown@placeholder.org	14FEB2025	460	455	24
21	EB020	Globalcornea	Singapore	APAC	global@cornea.org	15MAR2025	390	370	19
22	EB021	Brightvision	India	APAC	bright@vision.org	01APR2025	440	435	33
23	EB022	Cleargift	Uk	EMEA	cleargift@eye.org	18MAY2025	480	470	27

Explanation

This workflow combines multiple defensive programming approaches used in production clinical environments. PROPCASE standardizes presentation layers while STRIP removes hidden whitespace corruption. ABS corrects impossible negative operational values without destroying historical traceability. INTNX generates replacement dates using reproducible business rules. INDEX validates email structure quickly without expensive regular expressions. SELECT-WHEN logic improves readability compared with nested IF blocks and simplifies maintenance during production deployment.

3.Deduplication Using PROC SORT

proc sort data=eyebank_clean out=audit_reference nodupkey;

by eye_bank_id;

run;

proc print data=audit_reference;

run;

OUTPUT:

NOTE: There were 23 observations read from the data set WORK.EYEBANK_CLEAN.

NOTE: 1 observations with duplicate key values were deleted.

NOTE: The data set WORK.AUDIT_REFERENCE has 22 observations and 9 variables.

OUTPUT:

Obs	eye_bank_id	eye_bank_name	country	region_code	contact_email	last_audit_date	donors_received	transplants_done	avg_wait_days
1	EB001	Lionseyebank	India	APAC	support@lions.org	12JAN2025	450	420	32
2	EB002	Visiontrust	Usa	NA	info@visiontrust.org	15FEB2025	620	590	28
3	EB003	Globaleye	Uk	EMEA	admin@globaleye.org	11MAR2025	45	45	20
4	EB004	Corneacare	India	APAC	unknown@placeholder.org	04APR2026	520	520	18
5	EB005	Brightsight	Canada	NA	help@brightsight	14APR2025	410	390	25
6	EB006	Futurevision	Australia	APAC	unknown@placeholder.org	04APR2026	300	280	22
7	EB007	Eyelife	Germany	EMEA	contact@eyelife.org	19MAY2025	290	275	.
8	EB008	Worldcornea	India	APAC	world@cornea.org	04APR2026	610	605	31
9	EB009	Clearview	Japan	APAC	clear@view.org	10JUN2025	510	490	27
10	EB010	Visiontrust	Usa	NA	info@visiontrust.org	15FEB2025	620	590	28
11	EB011	Eye Hope	Brazil	LATAM	hope@eye.org	01JUL2025	400	400	30
12	EB012	Retinagift	India	APAC	retina@gift.org	20AUG2025	.	.	40
13	EB013	Newlight	Uae	MEA	newlight@mail.com	15SEP2025	370	355	29
14	EB014	Ocularbank	France	EMEA	ocular@bank.org	15OCT2025	280	270	23
15	EB015	Donorvision	India	APAC	donor@vision.org	22NOV2025	530	520	26
16	EB016	Lifesight	Kenya	AFRICA	sight@life.org	11DEC2025	250	240	34
17	EB017	Giftvision	India	APAC	gift@vision.org	04APR2026	0	0	36
18	EB018	Hopecornea	Usa	NA	hope@cornea.org	15JAN2025	550	530	21
19	EB019	Eyefuture	India	APAC	unknown@placeholder.org	14FEB2025	460	455	24
20	EB020	Globalcornea	Singapore	APAC	global@cornea.org	15MAR2025	390	370	19
21	EB021	Brightvision	India	APAC	bright@vision.org	01APR2025	440	435	33
22	EB022	Cleargift	Uk	EMEA	cleargift@eye.org	18MAY2025	480	470	27

Explanation

PROC SORT with NODUPKEY removes duplicate primary identifiers while preserving first occurrence records. This method is extremely common in SDTM and ADaM preparation pipelines. Analysts must document whether first occurrence retention aligns with business rules because inappropriate deduplication can create regulatory traceability concerns.

4.DATA Step Versus PROC SQL

proc sort data=eyebank_clean;

by eye_bank_id;

run;

proc print data=eyebank_clean;

run;

OUTPUT:

Obs	eye_bank_id	eye_bank_name	country	region_code	contact_email	last_audit_date	donors_received	transplants_done	avg_wait_days
1	EB001	Lionseyebank	India	APAC	support@lions.org	12JAN2025	450	420	32
2	EB002	Visiontrust	Usa	NA	info@visiontrust.org	15FEB2025	620	590	28
3	EB003	Globaleye	Uk	EMEA	admin@globaleye.org	11MAR2025	45	45	20
4	EB003	Globaleye	Uk	EMEA	admin@globaleye.org	11MAR2025	500	310	20
5	EB004	Corneacare	India	APAC	unknown@placeholder.org	04APR2026	520	520	18
6	EB005	Brightsight	Canada	NA	help@brightsight	14APR2025	410	390	25
7	EB006	Futurevision	Australia	APAC	unknown@placeholder.org	04APR2026	300	280	22
8	EB007	Eyelife	Germany	EMEA	contact@eyelife.org	19MAY2025	290	275	.
9	EB008	Worldcornea	India	APAC	world@cornea.org	04APR2026	610	605	31
10	EB009	Clearview	Japan	APAC	clear@view.org	10JUN2025	510	490	27
11	EB010	Visiontrust	Usa	NA	info@visiontrust.org	15FEB2025	620	590	28
12	EB011	Eye Hope	Brazil	LATAM	hope@eye.org	01JUL2025	400	400	30
13	EB012	Retinagift	India	APAC	retina@gift.org	20AUG2025	.	.	40
14	EB013	Newlight	Uae	MEA	newlight@mail.com	15SEP2025	370	355	29
15	EB014	Ocularbank	France	EMEA	ocular@bank.org	15OCT2025	280	270	23
16	EB015	Donorvision	India	APAC	donor@vision.org	22NOV2025	530	520	26
17	EB016	Lifesight	Kenya	AFRICA	sight@life.org	11DEC2025	250	240	34
18	EB017	Giftvision	India	APAC	gift@vision.org	04APR2026	0	0	36
19	EB018	Hopecornea	Usa	NA	hope@cornea.org	15JAN2025	550	530	21
20	EB019	Eyefuture	India	APAC	unknown@placeholder.org	14FEB2025	460	455	24
21	EB020	Globalcornea	Singapore	APAC	global@cornea.org	15MAR2025	390	370	19
22	EB021	Brightvision	India	APAC	bright@vision.org	01APR2025	440	435	33
23	EB022	Cleargift	Uk	EMEA	cleargift@eye.org	18MAY2025	480	470	27

4.1 DATA Step Merge

data audit_reference;

length eye_bank_id $8

       audit_status $20;

input eye_bank_id $ audit_status $;

datalines;

EB001 PASS

EB002 PASS

EB003 REVIEW

EB004 FAIL

EB005 PASS

EB006 REVIEW

EB007 PASS

EB008 REVIEW

EB009 PASS

EB010 PASS

EB011 FAIL

EB012 REVIEW

EB013 PASS

EB014 PASS

EB015 PASS

EB016 REVIEW

EB017 FAIL

EB018 PASS

EB019 REVIEW

EB020 PASS

EB021 PASS

EB022 PASS

;

run;

proc print data=audit_reference;

run;

OUTPUT:

Obs	eye_bank_id	audit_status
1	EB001	PASS
2	EB002	PASS
3	EB003	REVIEW
4	EB004	FAIL
5	EB005	PASS
6	EB006	REVIEW
7	EB007	PASS
8	EB008	REVIEW
9	EB009	PASS
10	EB010	PASS
11	EB011	FAIL
12	EB012	REVIEW
13	EB013	PASS
14	EB014	PASS
15	EB015	PASS
16	EB016	REVIEW
17	EB017	FAIL
18	EB018	PASS
19	EB019	REVIEW
20	EB020	PASS
21	EB021	PASS
22	EB022	PASS

data final_ds;

merge eyebank_clean(in=a)

      audit_reference(in=b);

by eye_bank_id;

if a;

run;

proc print data=final_ds;

run;

OUTPUT:

Obs	eye_bank_id	eye_bank_name	country	region_code	contact_email	last_audit_date	donors_received	transplants_done	avg_wait_days	audit_status
1	EB001	Lionseyebank	India	APAC	support@lions.org	12JAN2025	450	420	32	PASS
2	EB002	Visiontrust	Usa	NA	info@visiontrust.org	15FEB2025	620	590	28	PASS
3	EB003	Globaleye	Uk	EMEA	admin@globaleye.org	11MAR2025	45	45	20	REVIEW
4	EB003	Globaleye	Uk	EMEA	admin@globaleye.org	11MAR2025	500	310	20	REVIEW
5	EB004	Corneacare	India	APAC	unknown@placeholder.org	04APR2026	520	520	18	FAIL
6	EB005	Brightsight	Canada	NA	help@brightsight	14APR2025	410	390	25	PASS
7	EB006	Futurevision	Australia	APAC	unknown@placeholder.org	04APR2026	300	280	22	REVIEW
8	EB007	Eyelife	Germany	EMEA	contact@eyelife.org	19MAY2025	290	275	.	PASS
9	EB008	Worldcornea	India	APAC	world@cornea.org	04APR2026	610	605	31	REVIEW
10	EB009	Clearview	Japan	APAC	clear@view.org	10JUN2025	510	490	27	PASS
11	EB010	Visiontrust	Usa	NA	info@visiontrust.org	15FEB2025	620	590	28	PASS
12	EB011	Eye Hope	Brazil	LATAM	hope@eye.org	01JUL2025	400	400	30	FAIL
13	EB012	Retinagift	India	APAC	retina@gift.org	20AUG2025	.	.	40	REVIEW
14	EB013	Newlight	Uae	MEA	newlight@mail.com	15SEP2025	370	355	29	PASS
15	EB014	Ocularbank	France	EMEA	ocular@bank.org	15OCT2025	280	270	23	PASS
16	EB015	Donorvision	India	APAC	donor@vision.org	22NOV2025	530	520	26	PASS
17	EB016	Lifesight	Kenya	AFRICA	sight@life.org	11DEC2025	250	240	34	REVIEW
18	EB017	Giftvision	India	APAC	gift@vision.org	04APR2026	0	0	36	FAIL
19	EB018	Hopecornea	Usa	NA	hope@cornea.org	15JAN2025	550	530	21	PASS
20	EB019	Eyefuture	India	APAC	unknown@placeholder.org	14FEB2025	460	455	24	REVIEW
21	EB020	Globalcornea	Singapore	APAC	global@cornea.org	15MAR2025	390	370	19	PASS
22	EB021	Brightvision	India	APAC	bright@vision.org	01APR2025	440	435	33	PASS
23	EB022	Cleargift	Uk	EMEA	cleargift@eye.org	18MAY2025	480	470	27	PASS

4.2 PROC SQL Join

proc sql;

create table final_sql as

select a.*,

       b.audit_status

from eyebank_clean as a

left join audit_reference as b

on a.eye_bank_id=b.eye_bank_id;

quit;

proc print data=final_sql;

run;

OUTPUT:

Obs	eye_bank_id	eye_bank_name	country	region_code	contact_email	last_audit_date	donors_received	transplants_done	avg_wait_days	audit_status
1	EB001	Lionseyebank	India	APAC	support@lions.org	12JAN2025	450	420	32	PASS
2	EB002	Visiontrust	Usa	NA	info@visiontrust.org	15FEB2025	620	590	28	PASS
3	EB003	Globaleye	Uk	EMEA	admin@globaleye.org	11MAR2025	45	45	20	REVIEW
4	EB003	Globaleye	Uk	EMEA	admin@globaleye.org	11MAR2025	500	310	20	REVIEW
5	EB004	Corneacare	India	APAC	unknown@placeholder.org	04APR2026	520	520	18	FAIL
6	EB005	Brightsight	Canada	NA	help@brightsight	14APR2025	410	390	25	PASS
7	EB006	Futurevision	Australia	APAC	unknown@placeholder.org	04APR2026	300	280	22	REVIEW
8	EB007	Eyelife	Germany	EMEA	contact@eyelife.org	19MAY2025	290	275	.	PASS
9	EB008	Worldcornea	India	APAC	world@cornea.org	04APR2026	610	605	31	REVIEW
10	EB009	Clearview	Japan	APAC	clear@view.org	10JUN2025	510	490	27	PASS
11	EB010	Visiontrust	Usa	NA	info@visiontrust.org	15FEB2025	620	590	28	PASS
12	EB011	Eye Hope	Brazil	LATAM	hope@eye.org	01JUL2025	400	400	30	FAIL
13	EB012	Retinagift	India	APAC	retina@gift.org	20AUG2025	.	.	40	REVIEW
14	EB013	Newlight	Uae	MEA	newlight@mail.com	15SEP2025	370	355	29	PASS
15	EB014	Ocularbank	France	EMEA	ocular@bank.org	15OCT2025	280	270	23	PASS
16	EB015	Donorvision	India	APAC	donor@vision.org	22NOV2025	530	520	26	PASS
17	EB016	Lifesight	Kenya	AFRICA	sight@life.org	11DEC2025	250	240	34	REVIEW
18	EB017	Giftvision	India	APAC	gift@vision.org	04APR2026	0	0	36	FAIL
19	EB018	Hopecornea	Usa	NA	hope@cornea.org	15JAN2025	550	530	21	PASS
20	EB019	Eyefuture	India	APAC	unknown@placeholder.org	14FEB2025	460	455	24	REVIEW
21	EB020	Globalcornea	Singapore	APAC	global@cornea.org	15MAR2025	390	370	19	PASS
22	EB021	Brightvision	India	APAC	bright@vision.org	01APR2025	440	435	33	PASS
23	EB022	Cleargift	Uk	EMEA	cleargift@eye.org	18MAY2025	480	470	27	PASS

Comparison

DATA step merges are faster for sorted datasets and provide FIRST./LAST. functionality. PROC SQL offers superior readability for complex joins and aggregations. Enterprise teams typically use both depending on workflow requirements.

5.FIRST. and LAST. Processing

proc sort data=final_ds;

by country;

run;

proc print data=final_ds;

run;

OUTPUT:

Obs	eye_bank_id	eye_bank_name	country	region_code	contact_email	last_audit_date	donors_received	transplants_done	avg_wait_days	audit_status
1	EB006	Futurevision	Australia	APAC	unknown@placeholder.org	04APR2026	300	280	22	REVIEW
2	EB011	Eye Hope	Brazil	LATAM	hope@eye.org	01JUL2025	400	400	30	FAIL
3	EB005	Brightsight	Canada	NA	help@brightsight	14APR2025	410	390	25	PASS
4	EB014	Ocularbank	France	EMEA	ocular@bank.org	15OCT2025	280	270	23	PASS
5	EB007	Eyelife	Germany	EMEA	contact@eyelife.org	19MAY2025	290	275	.	PASS
6	EB001	Lionseyebank	India	APAC	support@lions.org	12JAN2025	450	420	32	PASS
7	EB004	Corneacare	India	APAC	unknown@placeholder.org	04APR2026	520	520	18	FAIL
8	EB008	Worldcornea	India	APAC	world@cornea.org	04APR2026	610	605	31	REVIEW
9	EB012	Retinagift	India	APAC	retina@gift.org	20AUG2025	.	.	40	REVIEW
10	EB015	Donorvision	India	APAC	donor@vision.org	22NOV2025	530	520	26	PASS
11	EB017	Giftvision	India	APAC	gift@vision.org	04APR2026	0	0	36	FAIL
12	EB019	Eyefuture	India	APAC	unknown@placeholder.org	14FEB2025	460	455	24	REVIEW
13	EB021	Brightvision	India	APAC	bright@vision.org	01APR2025	440	435	33	PASS
14	EB009	Clearview	Japan	APAC	clear@view.org	10JUN2025	510	490	27	PASS
15	EB016	Lifesight	Kenya	AFRICA	sight@life.org	11DEC2025	250	240	34	REVIEW
16	EB020	Globalcornea	Singapore	APAC	global@cornea.org	15MAR2025	390	370	19	PASS
17	EB013	Newlight	Uae	MEA	newlight@mail.com	15SEP2025	370	355	29	PASS
18	EB003	Globaleye	Uk	EMEA	admin@globaleye.org	11MAR2025	45	45	20	REVIEW
19	EB003	Globaleye	Uk	EMEA	admin@globaleye.org	11MAR2025	500	310	20	REVIEW
20	EB022	Cleargift	Uk	EMEA	cleargift@eye.org	18MAY2025	480	470	27	PASS
21	EB002	Visiontrust	Usa	NA	info@visiontrust.org	15FEB2025	620	590	28	PASS
22	EB010	Visiontrust	Usa	NA	info@visiontrust.org	15FEB2025	620	590	28	PASS
23	EB018	Hopecornea	Usa	NA	hope@cornea.org	15JAN2025	550	530	21	PASS

data country_summary;

set final_ds;

by country;

retain total_donors 0;

if first.country then total_donors=0;

total_donors+donors_received;

if last.country then output;

run;

proc print data=country_summary;

run;

OUTPUT:

Obs	eye_bank_id	eye_bank_name	country	region_code	contact_email	last_audit_date	donors_received	transplants_done	avg_wait_days	audit_status	total_donors
1	EB006	Futurevision	Australia	APAC	unknown@placeholder.org	04APR2026	300	280	22	REVIEW	300
2	EB011	Eye Hope	Brazil	LATAM	hope@eye.org	01JUL2025	400	400	30	FAIL	400
3	EB005	Brightsight	Canada	NA	help@brightsight	14APR2025	410	390	25	PASS	410
4	EB014	Ocularbank	France	EMEA	ocular@bank.org	15OCT2025	280	270	23	PASS	280
5	EB007	Eyelife	Germany	EMEA	contact@eyelife.org	19MAY2025	290	275	.	PASS	290
6	EB021	Brightvision	India	APAC	bright@vision.org	01APR2025	440	435	33	PASS	3010
7	EB009	Clearview	Japan	APAC	clear@view.org	10JUN2025	510	490	27	PASS	510
8	EB016	Lifesight	Kenya	AFRICA	sight@life.org	11DEC2025	250	240	34	REVIEW	250
9	EB020	Globalcornea	Singapore	APAC	global@cornea.org	15MAR2025	390	370	19	PASS	390
10	EB013	Newlight	Uae	MEA	newlight@mail.com	15SEP2025	370	355	29	PASS	370
11	EB022	Cleargift	Uk	EMEA	cleargift@eye.org	18MAY2025	480	470	27	PASS	1025
12	EB018	Hopecornea	Usa	NA	hope@cornea.org	15JAN2025	550	530	21	PASS	1790

Explanation

FIRST. and LAST. variables are among the most powerful SAS capabilities unavailable in standard SQL implementations. These automatic indicators support cumulative calculations, patient visit sequencing, exposure duration calculations, and longitudinal clinical analysis.

6.ARRAY Processing Example

data num_arrays;

set final_ds;

array nums donors_received

           transplants_done

           avg_wait_days;

do i=1 to dim(nums);

if nums(i)<0 then nums(i)=abs(nums(i));

end;

run;

proc print data=num_arrays;

run;

OUTPUT:

Obs	eye_bank_id	eye_bank_name	country	region_code	contact_email	last_audit_date	donors_received	transplants_done	avg_wait_days	audit_status	i
1	EB006	Futurevision	Australia	APAC	unknown@placeholder.org	04APR2026	300	280	22	REVIEW	4
2	EB011	Eye Hope	Brazil	LATAM	hope@eye.org	01JUL2025	400	400	30	FAIL	4
3	EB005	Brightsight	Canada	NA	help@brightsight	14APR2025	410	390	25	PASS	4
4	EB014	Ocularbank	France	EMEA	ocular@bank.org	15OCT2025	280	270	23	PASS	4
5	EB007	Eyelife	Germany	EMEA	contact@eyelife.org	19MAY2025	290	275	.	PASS	4
6	EB001	Lionseyebank	India	APAC	support@lions.org	12JAN2025	450	420	32	PASS	4
7	EB004	Corneacare	India	APAC	unknown@placeholder.org	04APR2026	520	520	18	FAIL	4
8	EB008	Worldcornea	India	APAC	world@cornea.org	04APR2026	610	605	31	REVIEW	4
9	EB012	Retinagift	India	APAC	retina@gift.org	20AUG2025	.	.	40	REVIEW	4
10	EB015	Donorvision	India	APAC	donor@vision.org	22NOV2025	530	520	26	PASS	4
11	EB017	Giftvision	India	APAC	gift@vision.org	04APR2026	0	0	36	FAIL	4
12	EB019	Eyefuture	India	APAC	unknown@placeholder.org	14FEB2025	460	455	24	REVIEW	4
13	EB021	Brightvision	India	APAC	bright@vision.org	01APR2025	440	435	33	PASS	4
14	EB009	Clearview	Japan	APAC	clear@view.org	10JUN2025	510	490	27	PASS	4
15	EB016	Lifesight	Kenya	AFRICA	sight@life.org	11DEC2025	250	240	34	REVIEW	4
16	EB020	Globalcornea	Singapore	APAC	global@cornea.org	15MAR2025	390	370	19	PASS	4
17	EB013	Newlight	Uae	MEA	newlight@mail.com	15SEP2025	370	355	29	PASS	4
18	EB003	Globaleye	Uk	EMEA	admin@globaleye.org	11MAR2025	45	45	20	REVIEW	4
19	EB003	Globaleye	Uk	EMEA	admin@globaleye.org	11MAR2025	500	310	20	REVIEW	4
20	EB022	Cleargift	Uk	EMEA	cleargift@eye.org	18MAY2025	480	470	27	PASS	4
21	EB002	Visiontrust	Usa	NA	info@visiontrust.org	15FEB2025	620	590	28	PASS	4
22	EB010	Visiontrust	Usa	NA	info@visiontrust.org	15FEB2025	620	590	28	PASS	4
23	EB018	Hopecornea	Usa	NA	hope@cornea.org	15JAN2025	550	530	21	PASS	4

Explanation

ARRAY processing reduces repetitive code and improves maintainability. In production SDTM environments, arrays often validate hundreds of laboratory parameters simultaneously.

7.Professional Reporting

proc report data=eyebank_clean nowd;

column country donors_received

       transplants_done

       avg_wait_days;

define country/group;

define donors_received/sum;

define transplants_done/sum;

define avg_wait_days/mean;

run;

OUTPUT:

country	donors_received	transplants_done	avg_wait_days
Australia	300	280	22
Brazil	400	400	30
Canada	410	390	25
France	280	270	23
Germany	290	275	.
India	3010	2955	30
Japan	510	490	27
Kenya	250	240	34
Singapore	390	370	19
Uae	370	355	29
Uk	1025	825	22.333333
Usa	1790	1710	25.666667

Explanation

PROC REPORT remains one of the most powerful reporting procedures in SAS. Clinical TLF generation frequently relies on REPORT because of its flexible compute blocks and formatting options.

8.Reusable Macro Example

%macro check_missing(ds,var);

proc sql;

select count(*) as missing_count

from &ds

where missing(&var);

quit;

%mend;

%check_missing(eyebank_clean,contact_email);

OUTPUT:

missing_count
0

Explanation

Macros improve standardization and reproducibility. Regulatory organizations strongly prefer reusable validated components rather than duplicated business logic.

9.R Raw Dataset

library(tibble)

eyebank_raw <- tibble(

  eye_bank_id = c("EB001","EB002","EB003","EB003","EB004","EB005","EB006",

    "EB007","EB008","EB009","EB010","EB011","EB012","EB013","EB014",

    "EB015","EB016","EB017","EB018","EB019","EB020","EB021","EB022"),

  eye_bank_name = c("LionsEyeBank","VisionTrust","GlobalEye","GlobalEye",

    "CorneaCare","BrightSight","FutureVision","EyeLife","WorldCornea",

    "ClearView","VisionTrust","Eye Hope","RetinaGift","NewLight",

    "OcularBank","DonorVision","LifeSight","GiftVision","HopeCornea",

    "EyeFuture","GlobalCornea","BrightVision","ClearGift"),

  country = c("India","USA","UK","UK","India","Canada","Australia",

    "Germany","India","Japan","USA","Brazil","India","UAE","France",

    "India","Kenya","India","USA","India","Singapore","India","UK"),

  region_code = c("APAC","NA","EUROPE","Europe","APAC","na","APAC",

    "EU","AsiaPacific","APAC","NA","LATAM","APAC","MEA","eu","APAC",

    "AFRICA","APAC","NorthAmerica","APAC","APAC","APAC","EUROPE"),

  donors_received = c(450,620,-45,500,520,410,300,290,610,510,620,400,

                      NA,370,280,530,250,0,550,460,390,440,480),

  transplants_done = c(420,590,310,310,700,390,280,275,605,490,590,410,

                       340,355,270,520,240,230,530,455,370,435,470),

  avg_wait_days = c(32,28,20,20,18,25,22,-5,31,27,28,30,40,29,

                    23,26,34,36,21,24,19,33,27),

  contact_email = c("support@lions.org","info@visiontrust.org",

    "admin@globaleye.org","admin@globaleye.org","NULL","help@brightsight",

    "futurevision.org","contact@eyelife.org","world@cornea.org","clear@view.org",

    "info@visiontrust.org","hope @eye.org","retina@gift.org","newlight@mail.com",

    "ocular@bank.org","donor@vision.org","sight@life.org","gift@vision.org",

    "hope@cornea.org","eyefuture.com","global@cornea.org","bright@vision.org",

    "cleargift@eye.org"),

  last_audit_date = c("12JAN2025","15FEB2025","11MAR2025","11MAR2025",".",

    "14APR2025","30FEB2025","19MAY2025",".","10JUN2025","15FEB2025","01JUL2025",

    "20AUG2025","15SEP2025","15OCT2025","22NOV2025","11DEC2025",".","15JAN2025",

    "14FEB2025","15MAR2025","01APR2025","18MAY2025")

)

OUTPUT:

eye_bank_id	eye_bank_name	country	region_code	donors_received	transplants_done	avg_wait_days	contact_email	last_audit_date
EB001	LionsEyeBank	India	APAC	450	420	32	support@lions.org	12JAN2025
EB002	VisionTrust	USA	NA	620	590	28	info@visiontrust.org	15FEB2025
EB003	GlobalEye	UK	EUROPE	-45	310	20	admin@globaleye.org	11MAR2025
EB003	GlobalEye	UK	Europe	500	310	20	admin@globaleye.org	11MAR2025
EB004	CorneaCare	India	APAC	520	700	18	NULL	.
EB005	BrightSight	Canada	na	410	390	25	help@brightsight	14APR2025
EB006	FutureVision	Australia	APAC	300	280	22	futurevision.org	30FEB2025
EB007	EyeLife	Germany	EU	290	275	-5	contact@eyelife.org	19MAY2025
EB008	WorldCornea	India	AsiaPacific	610	605	31	world@cornea.org	.
EB009	ClearView	Japan	APAC	510	490	27	clear@view.org	10JUN2025
EB010	VisionTrust	USA	NA	620	590	28	info@visiontrust.org	15FEB2025
EB011	Eye Hope	Brazil	LATAM	400	410	30	hope @eye.org	01JUL2025
EB012	RetinaGift	India	APAC		340	40	retina@gift.org	20AUG2025
EB013	NewLight	UAE	MEA	370	355	29	newlight@mail.com	15SEP2025
EB014	OcularBank	France	eu	280	270	23	ocular@bank.org	15OCT2025
EB015	DonorVision	India	APAC	530	520	26	donor@vision.org	22NOV2025
EB016	LifeSight	Kenya	AFRICA	250	240	34	sight@life.org	11DEC2025
EB017	GiftVision	India	APAC	0	230	36	gift@vision.org	.
EB018	HopeCornea	USA	NorthAmerica	550	530	21	hope@cornea.org	15JAN2025
EB019	EyeFuture	India	APAC	460	455	24	eyefuture.com	14FEB2025
EB020	GlobalCornea	Singapore	APAC	390	370	19	global@cornea.org	15MAR2025
EB021	BrightVision	India	APAC	440	435	33	bright@vision.org	01APR2025
EB022	ClearGift	UK	EUROPE	480	470	27	cleargift@eye.org	18MAY2025

10.R Cleaning Layer

library(tidyverse)

library(janitor)

library(lubridate)

eyebank_clean <- eyebank_raw %>%

  clean_names() %>%

  mutate(eye_bank_name=str_to_title(str_trim(eye_bank_name)),

               country=str_to_title(country),

           region_code=str_to_upper(region_code),

       donors_received=abs(donors_received),

         avg_wait_days=if_else(avg_wait_days<0,NA_real_,

                          avg_wait_days),

         contact_email=str_replace_all(contact_email," ",""),

         contact_email=coalesce(contact_email,"unknown@placeholder.org")

  ) %>%

  distinct(eye_bank_id,.keep_all=TRUE)

OUTPUT:

eye_bank_id	eye_bank_name	country	region_code	donors_received	transplants_done	avg_wait_days	contact_email	last_audit_date
EB001	Lionseyebank	India	APAC	450	420	32	support@lions.org	12JAN2025
EB002	Visiontrust	Usa	NA	620	590	28	info@visiontrust.org	15FEB2025
EB003	Globaleye	Uk	EUROPE	45	310	20	admin@globaleye.org	11MAR2025
EB004	Corneacare	India	APAC	520	700	18	NULL	.
EB005	Brightsight	Canada	NA	410	390	25	help@brightsight	14APR2025
EB006	Futurevision	Australia	APAC	300	280	22	futurevision.org	30FEB2025
EB007	Eyelife	Germany	EU	290	275		contact@eyelife.org	19MAY2025
EB008	Worldcornea	India	ASIAPACIFIC	610	605	31	world@cornea.org	.
EB009	Clearview	Japan	APAC	510	490	27	clear@view.org	10JUN2025
EB010	Visiontrust	Usa	NA	620	590	28	info@visiontrust.org	15FEB2025
EB011	Eye Hope	Brazil	LATAM	400	410	30	hope@eye.org	01JUL2025
EB012	Retinagift	India	APAC		340	40	retina@gift.org	20AUG2025
EB013	Newlight	Uae	MEA	370	355	29	newlight@mail.com	15SEP2025
EB014	Ocularbank	France	EU	280	270	23	ocular@bank.org	15OCT2025
EB015	Donorvision	India	APAC	530	520	26	donor@vision.org	22NOV2025
EB016	Lifesight	Kenya	AFRICA	250	240	34	sight@life.org	11DEC2025
EB017	Giftvision	India	APAC	0	230	36	gift@vision.org	.
EB018	Hopecornea	Usa	NORTHAMERICA	550	530	21	hope@cornea.org	15JAN2025
EB019	Eyefuture	India	APAC	460	455	24	eyefuture.com	14FEB2025
EB020	Globalcornea	Singapore	APAC	390	370	19	global@cornea.org	15MAR2025
EB021	Brightvision	India	APAC	440	435	33	bright@vision.org	01APR2025
EB022	Cleargift	Uk	EUROPE	480	470	27	cleargift@eye.org	18MAY2025

Explanation

The tidyverse workflow mirrors SAS logic almost line for line. mutate() behaves similarly to DATA step assignments while case_when() resembles SELECT-WHEN blocks. distinct() mirrors PROC SORT NODUPKEY behavior and coalesce() provides elegant missing value replacement.

Enterprise Validation and Compliance

Clinical data cleaning extends beyond aesthetics.

It supports:

• SDTM traceability
• ADaM reproducibility
• Audit trail requirements
• QC independence
• Regulatory transparency
• Submission readiness

One dangerous SAS behavior involves missing numeric values.

In SAS:

if age<18 then pediatric='Y';

Missing ages satisfy this condition because missing numeric values are treated as smaller than valid numbers.

This subtle behavior has caused real production errors.

Business Logic Behind Cleaning Decisions

Data cleaning is never arbitrary.

Negative donor counts become absolute values because donor registrations cannot be negative in reality. Missing audit dates are imputed using operational assumptions so monitoring reports continue functioning. Region codes are standardized because APAC and AsiaPacific represent identical business concepts but fragment aggregation logic.

Suppose one eye bank reports donor count as -450 due to accidental keyboard entry. Leaving this value unchanged could incorrectly reduce global donor estimates by hundreds of procedures.

Similarly, missing dates can break time-series analysis and trend forecasting.

Text normalization improves grouping consistency.

Email validation supports operational communication workflows.

Healthcare analytics depends on semantic consistency as much as statistical correctness.

Twenty Data Cleaning Best Practices

1. Preserve raw data permanently.
2. Never overwrite source files.
3. Validate identifiers first.
4. Standardize metadata.
5. Maintain audit logs.
6. Document assumptions.
7. Version control macros.
8. Separate QC teams.
9. Validate dates early.
10. Normalize text variables.
11. Track imputation rules.
12. Use reusable functions.
13. Automate validation reports.
14. Compare frequencies before deployment.
15. Protect lineage information.
16. Monitor truncation risk.
17. Standardize missing values.
18. Perform peer review.
19. Validate business rules.
20. Maintain reproducible workflows.

Twenty One-Line Insights

• Dirty data creates expensive business mistakes.
• Validation logic beats visual inspection.
• Metadata drives reproducibility.
• Missing values deserve business context.
• Duplicates distort reality.
• Standardization improves trust.
• Defensive programming saves projects.
• Clean inputs create reliable AI.
• Audits reward traceability.
• Governance reduces surprises.
• Business rules matter.
• Automation reduces human error.
• Documentation improves continuity.
• SQL is powerful but not magical.
• DATA step remains irreplaceable.
• Reusable macros reduce risk.
• Consistency improves analytics.
• Data lineage protects organizations.
• Regulatory trust starts with quality.
• Reliable decisions require reliable data.

SAS Versus R for Enterprise Cleaning

SAS provides unmatched auditability, metadata management, and production deployment reliability. DATA step programming remains exceptionally efficient for row-wise transformations and longitudinal clinical processing.

R provides flexibility, open-source innovation, and advanced string manipulation capabilities. Tidyverse workflows are highly expressive and often easier for exploratory analytics.

Large pharmaceutical organizations continue using SAS because regulatory ecosystems depend heavily on traceability and reproducibility.

Technology companies often prefer R because of integration with machine learning ecosystems.

The strongest analytical teams use both.

SAS handles governance.

R handles flexibility.

Together they create robust analytical intelligence.

Conclusion

Modern analytics systems depend less on algorithms and more on trustworthy input data.

An advanced predictive model trained on corrupted information simply produces sophisticated errors.

Eye bank operations illustrate this challenge clearly.

Duplicate identifiers distort operational capacity estimates. Missing audit dates interrupt compliance monitoring. Invalid email structures disrupt communication systems. Inconsistent regional classifications create misleading dashboards.

Data cleaning therefore becomes a strategic capability rather than an operational inconvenience.

SAS contributes industrial-grade governance, reproducibility, and audit readiness.

R contributes agility, innovation, and exploratory flexibility.

Organizations that combine both ecosystems gain significant competitive advantages in healthcare analytics, insurance risk modeling, financial fraud detection, and retail intelligence.

The future of analytics does not belong to the organization with the largest data warehouse.

It belongs to the organization with the most trustworthy data.

Because in modern analytics, vision alone is not enough.

Validated vision changes decisions.

Validated decisions change outcomes.

Interview Questions and Answers

1. How would you identify duplicate records in SAS?

Use PROC SORT with NODUPKEY or PROC SQL GROUP BY with HAVING COUNT(*)>1 depending on reporting requirements.

2. Why is LENGTH placement important in SAS?

LENGTH statements must appear before assignments because variable storage allocation occurs during compilation and late declarations create truncation risk.

3. What is the R equivalent of PROC SORT NODUPKEY?

The equivalent approach is distinct() from dplyr.

4. Why are missing values dangerous in SAS?

Missing numeric values are treated as smaller than valid numbers, causing unexpected IF condition behavior.

5. When would you choose PROC SQL over DATA Step?

PROC SQL performs better for complex joins and aggregations while DATA step excels at sequential processing and FIRST./LAST. logic.

--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

About the Author:

SAS Learning Hub is a data analytics and SAS programming platform focused on clinical, financial, and real-world data analysis. The content is created by professionals with academic training in Pharmaceutics and hands-on experience in Base SAS, PROC SQL, Macros, SDTM, and ADaM, providing practical and industry-relevant SAS learning resources.

Disclaimer:

The datasets and analysis in this article are created for educational and demonstration purposes only. Here we learn about EYE BANK DATA.

Our Mission:

This blog provides industry-focused SAS programming tutorials and analytics projects covering finance, healthcare, and technology.

This project is suitable for:

·  Students learning SAS

·  Data analysts building portfolios

·  Professionals preparing for SAS interviews

·  Bloggers writing about analytics and smart cities

--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Follow Us On : 

 

--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

--->Follow our blog for more SAS-based analytics projects and industry data models.

---> Support Us By Following Our Blog..

To deepen your understanding of SAS analytics, please refer to our other data science and industry-focused projects listed below:

1.How Can SAS Map The Journey And Influence Of India’s Freedom Fighters?

2.Can SAS Identify The Best Project And Tool Combinations For Software Teams?

3.Can SAS Reveal Which Office Environments Are Truly Productive And Which Are Not?

--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

About Us | Contact | Privacy Policy